Thesis (MPhil)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: There is more to language than just its formal structural properties and, similarly, more to language function than just its communicative and naming function. Language does not exist independent of society. As a part of society, it is used in a diversity of functions: it influences thought processes, constitutes what people perceive as reality, and produces, reproduces and denies prejudices. It is in pursuit of its ideological function that language plays a significant role in the establishment and maintenance of systematically asymmetrical power relations. This study focuses on the role that language plays in efforts to position the African National Congress (ANC) as more fit to govern than other political parties in South Africa. Adopting a qualitative research strategy, the study provides an analysis of the discourse that is constructed in the ANC’s 1999, 2004 and 2009 national election manifestos. The analysis is presented within the framework of Critical Discourse Analysis (CDA) and is performed in terms of linguistic devices, techniques and strategies such as genre and its sequential structure, pronouns, contrasting expressions, intertextuality, grounding and elision, statistics and numerical figures, and discourse. It is demonstrated that the three election manifestos are situated within a specific socio-economic and political context defined by poverty, unemployment and inequality, which are rooted in the South African history of colonialism and race-based capitalism. The texts draw from resources of the genre of manifesto and show common structural features. It is shown that ambiguous pronouns are used to build up affinities between the ANC and the reader/listener with respect to the achievements of the ANC-led government, what work still needs to be done, and to position the ANC’s vision as one that is generally shared by the people. Contrasting expressions are used to disparage the apartheid system and to extol the post-1994 democratic system. In all three texts the ANC is foregrounded as the organization which not only brought freedom to South Africa, but which in fact led the struggle for freedom and change. At the same time, there is an omission of other political organizations and the role they played in this struggle. It is also demonstrated that the three texts constituted by elements of other texts such as the Freedom Charter (1955), the Reconstruction and Development Programme (1994) and the Constitution (1996) use statistics and figures to bestow the ANC with a systematic and scientific gravitas. Lastly, the three manifestos reflect a discourse of “complete” or “total” freedom, which is inclusive of the social, economic and political aspects of the reality of South Africans’ lives. It is argued that these linguistic devices, techniques and strategies are used in the 1999, 2004 and 2009 national election manifestos to position the ANC as more fit to govern South Africa than other political parties. / AFRIKAANSE OPSOMMING: Taal behels meer as net formele strukturele eienskappe, en die funksies van taal behels eweneens meer as net benaming en kommunikasie. Taal bestaan nie onafhanklik van die gemeenskap nie. As ’n deel van die gemeenskap, word taal in ’n verskeidenheid funksies gebruik: dit beïnvloed denkprosesse, bepaal wat mense beskou as die werklikheid, en dien om vooroordele te skep, te verhoog en te ontken. Dit is in die uitoefening van sy ideologiese funksie dat taal ’n beduidende rol speel in die vestiging en handhawing van sistematies asimmetriese magsverhoudings. Hierdie studie fokus op die rol wat taal speel in pogings om die African National Congress (ANC) te posisioneer as meer geskik om te regeer as ander politieke partye in Suid-Afrika. Met ’n kwalitatiewe navorsingstrategie as uitgangspunt, bied die studie ’n analise van die diskoers wat gekonstrueer word in die ANC se onderskeie manifeste vir die 1999, 2004 en 2009 nasionale verkiesings. Die analise word aangebied binne die raamwerk van Kritiese Diskoersanalise (“Critical Discourse Analysis”) en word uitgevoer in terme van taalkundige meganismes, tegnieke en strategieë soos genre and sy sekwensiële struktuur, voornaamwoorde, teenstellende uitdrukkings, intertekstualiteit, opstelling en weglating (“grounding and elision”), statistieke en getalle, en diskoers. Daar word aangetoon dat die drie verkiesingsmanifeste ingebed is in ’n spesifieke sosio-ekonomiese en politieke konteks van armoede, werkloosheid en ongelykheid, wat gegrond is in die Suid-Afrikaanse geskiedenis van kolonialisme en rasgebaseerde kapitalisme. Die tekste benut die middele van die manifes-genre en vertoon gemeenskaplike strukturele kenmerke. Daar word aangetoon hoe dubbelsinnige voornaam-woorde gebruik word om ’n affiniteit tussen die ANC en die leser/hoorder tot stand te bring ten opsigte van die ANC-regering se prestasies, die werk wat nog gedoen moet word, en ook om die ANC se visie voor te hou as een wat algemeen deur die mense gedeel word. Teenstellende uitdrukkings word gebruik om die apartheidstelsel te verdoem en die post-1994 demokratiese stelsel op te hemel. In al drie tekste word die ANC vooropgestel as die organisasie wat nie net vryheid na Suid-Afrika gebring het nie, maar wat in feite die stryd om vryheid en verandering gelei het. Terselfdertyd word geen melding gemaak van ander politieke organisasies en die rol wat hulle in dié stryd gespeel het nie. Daar word ook aangetoon dat die drie tekste wat verskeie elemente insluit van ander tekste soos die Freedom Charter (1955), die Heropbou- en Ontwikkelingsprogram (“Reconstruction and Development Programme”, 1994) en die Grondwet (1996) gebruik maak van statistieke en getalle om die ANC te bedeel met ’n sistematiese en wetenskaplike gravitas. Die drie manifeste vertoon, laastens, ’n diskoers van “totale” of “volledige” vryheid, wat die sosiale, ekonomiese en politieke aspekte van die werklikheid van Suid-Afrikaners se lewens omvat Daar word geargumenteer dat dié taalkundige meganismes, tegnieke en strategieë in die 1999, 2004 en 2009 verkiesingsmanifeste gebruik word om die ANC te posisioneer as meer geskik om te regeer as ander politieke partye.

Thesis (MPhil)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: The aim of this study is to evaluate the extent to which women in higher education experience or are aware of possible limitations placed on them as women in work contexts that historically were the domain of men only. The research critically analysed semistructured interview responses from seven female participants from a South African university. The analysis aimed to investigate the participants’ individual and shared perceptions and to present a view of how they discursively construct their own experience within an academic work environment. The research draws on theories and methodologies developed within critical discourse analysis (CDA) in order to interpret the data. This study draws on Gee’s model of CDA, which conceptualises discourse as inherently political and ideological, and therefore seeks to make clear the discursive connections between discourses and the sociocultural, historical and institutional contexts in which it is created and interpreted. Thus, a close analysis of discourses provides insight into aspects of social reality, including how individuals construct themselves and their personal experiences, as well as their attitudes and assumptions about their social contexts. The study reveals that during their narratives the women participants addressed similar issues and themes relating to gender in an academic work place. The participants also used similar linguistic and discursive strategies to construct their narratives. Subsequent to an analysis of the data using Gee’s model of CDA, it was found that the participants do experience limitations in the workplace, but they did not always directly attribute these limitations to gender. After a final analysis of the ways in which participants express attitudes to the issues being addressed, the study finds that the participants are not often aware of dominant ideologies related to gender, or of how the ideologies affect their experiences. / AFRIKAANSE OPSOMMING: Die doelwit van hierdie studie is om die mate waarin vroue in hoër onderwys ondervind of bewus is van moontlike beperkings wat op hulle geplaas word as vroue in werksomstandighede wat geskiedkundig deur mans domineer word. Semi-gestruktureerde onderhoude met sewe vroulike deelnemers by ʼn Suid-Afrikaanse universiteit word krities ontleed. Die ontleding is daarop gemik om die deelnemers se individuele en gedeelde waarnemings te ondersoek, asook om die diskursiewe vertelling van hul ervarings uit te beeld. Teorieë en metodologieë wat ontwikkel is in kritiese diskoersanalise (CDA) word gebruik om die data te interpreteer, terwyl daar spesifieke gebruik gemaak word van Gee se model van kritiese analise. Gee se model konseptualiseer diskoers as inherent polities en ideologies en poog daarom om die diskursiewe verbande uit te wys wat vorm tussen diskoerse en sosiokulturele, geskiedkundige en institusionele kontekste. Dus, ʼn noukeurige ontleding van diskoers bied insig tot aspekte van die sosiale werklikheid, insluitend hoe individue hulself en hul persoonlike ervarings, en hul houdings en aannames oor sosiale kontekste opbou. Die studie toon aan dat die vroulike deelnemers soortgelyke kwessies en temas, met betrekking tot geslag in ʼn akademiese werksplek, tydens hul onderhoude bespreek het. Die deelnemers het ook soortgelyke taalkundige en diskursiewe strategieë benut om hul verhale te struktureer. Na ʼn ontleding van die data met behulp van Gee se model van kritiese analise, is daar gevind dat die deelnemers wel beperkings in die werksplek ervaar, alhoewel hul die beperkings nie altyd direk aan geslag toegeskryf het nie. Na ʼn finale ontleding van die maniere waarop deelnemers hul houdings teenoor die kwessies uitbeeld, het die studie gevind dat die deelnemers dikwels nie bewus is van die dominante ideologieë wat verband hou met geslag nie, of hoe hierdie ideologieë hul ervarings beïnvloed nie.

Thesis (MA)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Children with specific learning disability (SLD) struggle with reading, spelling and/or writing, and tend to have poor narrative skills (Fey, Catts, Proctor-Williams, Tomblin & Zhang, 2004; Scanlon, 2013). Although learners with SLD mostly use grammatical sentences after discharge from conventional speech-language therapy, they often still struggle with creating narratives and find comprehension tasks challenging. For this study, I chose an action research design (McNiff & Whitehead, 2013). I used a lesser known therapy approach with some of these learners, in an attempt to ascertain whether kinetic methods of teaching language can improve their narrative skills. The research question posed was: What changes in narrative skills (if any), measured in terms of micro and macro structure, are brought about by the Sandtray Play Approach (STPA) in first language (L1) Afrikaans-speaking Grade 3 learners with SLDs? I used the STPA with four L1 Afrikaans-speaking Grade 3 boys with SLD over a course of six weeks (on average two therapy sessions per week). During each session, they each choose a variety of miniature objects with which they individually built a so-called “Sandworld” in their own sandtray. After each Sandworld was built, the boys took turns telling their story and listening to those of the other group member. After each narration, each listener made one positive comment and asked one question about the story. During the intervention period, I regularly presented mini-sessions teaching the participants how to create, improve and expand their narratives. The Language in Multilingual Society: Multilingual Assessment of Narratives (LITMUS: MAIN-Afrikaans) (Gagarina, Klop, Kunnari, Tantele, Valimaa, Baluciuniene, Bohnacker, & Walters, 2012) was used to analyse pre- and post-intervention narratives in terms of story structure components; (in)completeness of episodes; mean length of utterance (in words); and use of subordinating and co-ordinating conjunctions, conjunctive adverbs and internal state terms. The pre- and post-intervention results of the four boys in the experimental group were compared to those of four boys who did not receive any STPA intervention and who formed the control group. All eight participants had been clinically diagnosed as presenting with an SLD. Comparisons of the LITMUS: MAIN-Afrikaans scores indicated inter-group differences: Post-intervention, the experimental group used a higher number of story structure components, internal state terms, words, and utterances (those linguistic aspects introduced and practised during STPA intervention) than the control group. The experimental group’s Sandtray narratives also increased in multiple areas of structural complexity – e.g., in the use of conjunctive adverbs; subordinating and co-ordinating conjunctions; and embedding. These results have implications for speech-language therapy practice, showing that the STPA is potentially a valuable resource for teaching narrative skills to learners with SLD, and that positive changes can be observed in as little as six weeks. The STPA has been used in Europe with learners with hearing or learning impairment and with mainstream learners. In the South African context, it appears only to have been used to teach oral language use to deaf learners; and no South African studies on the use of the STPA for narrative skill development could be traced. This study was the first of its kind and indicated that using the STPA can assist Speech-Language Therapists in developing narrative skills of children with atypical language development. / AFRIKAANSE OPSOMMING: Kinders met spesifieke leergestremdheid (SLG) sukkel met lees, spelling en/of skryf, en neig om swak narratiefvaardighede te hê (Fey, Catts, Proctor-Williams, Tomblin & Zhang, 2004; Scanlon, 2013). Alhoewel leerders met SLG meestal grammatikale sinne gebruik ná ontslag uit konvensionele spraak-taalterapie, sukkel hulle steeds gereeld met narratiefproduksie en begripstake. Vir hierdie studie het ek gekies om ‘n aksienavorsingsmetode (McNiff & Whitehead, 2013) te gebruik. Ek het ‘n minder bekende terapiebenadering met sommige van hierdie leerders gevolg, in 'n poging om vas te stel of kinetiese metodes van taalonderrig hulle narratiefvaardighede kan verbeter. Die navorsingsvraag was: Watter veranderinge in narratiefvaardighede (indien enige), gemeet in terme van makro- en mikrostruktuur, word teweeg gebring deur die Sandbak-Speelbenadering (SBSB) in eerstetaal- (T1) Afrikaanssprekende Graad 3-leerders met SLG? Ek het die SBSB met vier T1 Afrikaanssprekende Graad 3-seuns met SLG oor ses weke gebruik (gemiddeld twee terapiesessies per week). Tydens hierdie sessies het elkeen ‘n verskeidenheid miniatuur voorwerpe gekies waarmee elkeen individueel ‘n sogenaamde “Sandwêreld” in hulle eie sandbakkie gebou het. Nadat elke Sandwêreld gebou is, het die seuns beurte geneem om hulle storie te vertel en na die stories van die ander groeplede te luister. Nadat elke storie vertel is, het elke luisteraar een positiewe punt van kommentaar oor die storie gelewer en een vraag oor die storie gevra. Gedurende die intervensietydperk het ek gereeld mini-sessies aangebied om die deelnemers te leer hoe om narratiewe te skep, te verbeter en uit te brei. Die Language in Multilingual Society: Multilingual Assessment of Narratives (LITMUS: MAIN-Afrikaans) (Gagarina, Klop, Kunnari, Tantele, Valimaa, Baluciuniene, Bohnacker, & Walters, 2012) is gebruik om voor- en ná-intervensie-narratiewe te ontleed in terme van storiestruktuurkomponente; (on)volledigheid van episodes; gemiddelde lengte van uiting (in woorde); en die gebruik van onderskikkende en neweskikkende voegwoorde, verbindingsbywoorde en interne toestand-terme. Ek het hierdie tellings vergelyk met dié behaal deur 'n kontrolegroep (vier manlike klasmaats van die eksperimentele groep). Ek het ook 'n seleksie van drie verteenwoordigende sandbaknarratiewe per eksperimentele groepdeelnemer in terme van hierdie metings geassesseer Al ag deelnemers is klinies gediagnoseer met SLG. Vergelykings van die LITMUS: MAIN-Afrikaans-tellings het inter-groepsverskille aangetoon: Die eksperimentele groep het ná intervensie ‘n groter aantal storiestruktuurkomponente, interne toestand-terme, woorde en uitinge (d.i. daardie linguistieke aspekte wat bekendgestel en geoefen is tydens SBSB-intervensie) gebruik as die kontrolegroep. Die eksperimentele groep se sandbaknarratiewe het ook toegeneem in terme van veelvuldige areas van strukturele kompleksiteit – byvoorbeeld in die gebruik van verbindingsbywoorde, onderskikkende en neweskikkende voegwoorde, en inbedding. Hierdie resultate het implikasies vir spraak-taalterapiepraktyk: Dit toon aan dat die SBSB potensieel ‘n waardevolle hulpbron is vir die leer van narratiefvaardighede aan leerders met SLG, en dat positiewe veranderinge in so min as ses weke waarneembaar is. Die SBSB word in Europa met leerders met gehoor- of leergestremdheid gebruik asook met hoofstroomleerders. In die Suid-Afrikaanse konteks is dit skynbaar nog slegs gebruik om gehoorgestremdes te leer praat; geen Suid-Afrikaanse studies oor die gebruik van die SBSB vir narratiefvaardigheidsontwikkeling kon gevind word nie. Hierdie studie was die eerste van sy soort en toon aan dat die gebruik van die SBSB Spraak-Taalterapeute kan help om narratiefvaardighede te ontwikkel in kinders met atipiese taalontwikkeling.

Thesis (MA)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: The decision by teachers to use a specific language for teaching in a bilingual context is influenced by a number of factors. These may include learners’ linguistic background, parental preferences on the use of language for teaching and learning, policy stipulations on language use, as well as learners’ cognitive level and their ability to comprehend lessons given in a specific language. Although policy stipulations and parental preference may emphasise the use of one particular language for teaching and learning, research shows that the classroom context and the dilemma teachers face in terms of language comprehension often play a role in the use of more than one language. Teachers often switch codes in a bilingual classroom for different reasons: clarifying subject matter, concept elaboration, encouraging leaners to participate, supporting exploratory talk, ensuring comprehension as well as switching codes as a classroom management strategy. This thesis investigates the frequency and the function of teachers’ code switching (CS) in a bilingual classroom context. The study was conducted in Limpopo Province, South Africa, in the Vhembe District Municipality. Two bilingual primary schools under Sibasa Circuit were chosen for study. A series of lessons were observed in the two schools and teachers’ interactions with learners in the classroom were recorded and then analysed qualitatively, guided by Myers- Scotton’s (1993) Markedness Model which is used to provide an account for different types of CS. This model is used to account for the motivations for every code choice in any discourse. The results of the study show that CS is a common feature in the two schools. The policy stipulations and parents’ preference do not limit teachers’ use of CS in such bilingual primary school classrooms. / AFRIKAANSE OPSOMMING: ’n Onderwyser se besluit om ’n spesifieke taal vir onderrig in ’n tweetalige konteks te gebruik word deur ’n reeks faktore beïnvloed. Dit sluit die leerders se taalagtergrond, ouers se taalvoorkeur vir onderrig en leer, amptelike skoolbeleid in verband met taalgebruik, sowel as die leerders se kognitiewe vlak en hul vermoë om klasse te verstaan wat in ’n spesifieke taal aangebied word, in. Hoewel skoolbeleid en ouers se taalvoorkeur die gebruik van een spesifieke taal vir onderrig en leer beklemtoon, toon navorsing dat die konteks van die klaskamer en die dilemma wat onderwysers in die gesig staar in terme van taalbegrip, ’n rol speel in die gebruik van meer as een taal. Onderwysers gebruik dikwels twee tale in ’n tweetalige klaskamer, om verskeie redes: verduideliking van lesmateriaal, uitbreiding van konsepte, aanmoediging van die leerders om deel te neem, ondersteuning van ondersoekende gesprekke, versekering van begrip sowel as kodewisseling as ’n strategie vir die bestuur van die klaskamer. Hierdie tesis ondersoek die frekwensie en die funksie van onderwysers se kodewisseling in die konteks van ’n tweetalige klaskamer. Die studie is in die Vhembe Distriksmunisipaliteit van die Limpopo Provinsie, Suid- Afrika, uitgevoer. Twee tweetalige laerskole wat deel uitmaak van die Sibasa-streek is gekies vir die studie. ’n Reeks klasse is waargeneem in die twee skole en die onderwysers se interaksies met die leerders in die klaskamer is opgeneem en daarna kwalitatief ontleed , gegrond op Myers- Scotton (1993) se Gemarkeerdheidsmodel (“Markedness Model”) wat gebruik word om ’n verklaring te gee vir die verskillende tipes kodewisseling. Hierdie model is gebruik om ’n verantwoording te bied van die redes vir elke kodekeuse in enige diskoers. Die resultate van die studie toon dat kodewisseling ’n algemene verskynsel in die twee skole is. Die skoolbeleid en ouers se taalvoorkeur beperk nie die onderwysers se gebruik van kodewisseling in sulke tweetalige laerskoolklaskamers nie. Onderwysers gebruik om verskeie redes kodewisseling, insluitend uitbreiding, verduideliking, beklemtoning, en teregwysing as ’n dissiplinêre strategie.

Thesis (MA)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: The aim of this study is to identify factors that affect the language choices and patterns of language use of multilingual speakers in the Western Cape, focusing particularly on emotional language. It is an exploratory study, taking a purely quantitative methodological approach. Data was collected by means of a web-based Multilingualism and Emotions Questionnaire available online for six months. The questionnaire was based on Dewaele and Pavlenko‟s Bilingualism and Emotions Questionnaire (see Pavlenko, 2005) and included multiple choice and Likert scale questions regarding participants‟ language use preferences, as well as their perceptions of each of their languages. Data analysis was split into two stages: firstly the response data was illustrated by means of frequency tables, and secondly the statistical method of Correspondence Analysis was used to show the patterns of variation among the factors investigated. Two hypotheses were proposed, based on previous research: firstly, that the sociolinguistic factors would play a bigger role than the socio-demographic factors in determining language choices and patterns of use, and secondly that there may be differences in patterns of use for speakers of African languages versus speakers of Western languages. The analysis confirmed both hypotheses, while also showing some deviation from the results of previous research, which is attributed to the context in which this study was conducted. The main findings of this study were that English was generally the preferred language even when not the L1, and also that Xhosa tended to follow an entirely different pattern of use in comparison with most other languages in the sample. These findings are attributed to the unique language contact situation in the Western Cape, showing that the widely accepted L1-primacy ideology does not quite hold true across all contexts. An English-bias in the implementation of the language and education policy is also identified as a possible contributing factor to the patterns of language use and language attitudes revealed in this study, hence it is suggested that methodological and practical changes to the language and education policy could lead to a realization of the true multilingual and multicultural potential of South Africa. / AFRIKAANSE OPSOMMING: Hierdie studie handel oor die identifisering van faktore wat die taalgebruikspatrone en keuses van veeltaliges in die Wes-Kaap affekteer. Die fokus is hoofsaaklik op emosionele taalgebruik. Die studie is verkennend, en volg „n suiwer kwantitatiewe benadering. Data is ingesamel deur die gebruik van „n web-gebaseerde Veeltaligheid en Emosies Vraelys wat ongeveer ses maande op die internet beskikbaar was. Die vraelys is gebaseer op dié van Dewaele en Pavlenko (verwys na Pavlenko, 2005) en sluit in veelvoudigekeuse- en Likertskaal-vrae oor die deelnemers se taalgebruik keuses asook hul taal-persepsies. Data-analise is in twee verskillende fases voltooi: eerstens is die data deur die gebruik van frekwensie tabelle geïllustreer, en tweedens is die patrone van variasie onder die faktore getoon deur die gebruik van die statistiese metode van Korrespondensie Ontleding. Twee hipoteses is voorgestel, gebaseer op vorige navorsing; eerstens, dat sosiolinguistiese faktore „n groter rol as sosio-demografiese faktore sou speel in die bepaling van taal-keuses en gebruikpatrone; en tweedens dat daar dalk verskille sou wees tussen Afrika-taal sprekers en Westerse-taal sprekers in verband met patrone van taalgebruik. Beide van die hipoteses is deur die analise bevestig, terwyl daar ook afwyking was van die bevindings van vorige navorsing, waarvoor die konteks waarin hierdie studie plaasvind as rede voorgestel is. Die studie se hoofbevindings dui daarop dat Engels oor die algemeen die voorgekeurde taal is, selfs wanneer dit nie die eerstetaal is nie, en ook dat Xhosa „n gebruikspatroon volg wat van die ander tale in die studie verskil. Hierdie bevindings word toegeskryf aan die uniekheid van die taal-kontak situasie in die Wes-Kaap, en dui daarop dat die algemeen aanvaarde ideologie van die eerste-taal-voorrang nie noodwendig van toepassing is op alle kontekste nie. „n Vooroordeel ten gunste van Engels in die implementering van die taal- en onderwysbeleid word ook uitgewys as „n moontlike bydraende faktor tot die patrone van taalgebruik en taalhoudings wat in hierdie studie ontdek is. Daar word gevolglik voorgestel dat metodologiese- en praktiese- veranderinge aan die taal- en onderwysbeleid groot bydrae sal kan lewer tot die vervulling van Suid-Afrika se volle veeltalige en multikulturele potensiaal.

The research reported in this thesis is basically applied in purpose. However the theoretical siqnificance of interactional sociolinquistics is explored by showinq that it is based on a philosophy of science which differs fundamentally from the versions of positivism which have informed linquistics over the years. The research methods consistent with this methodology are also outlined. The applied siqnificance of the sub-field is demonstrated more qenerally at first by examininq its contribution to the understandinq of the relationship between lanquaqe and context. Thereafter the contribution to the understandinq of this relationship is explored in more specific terms by examininq the role of contextual information in the form of culturally-specific interactional styles in the accomplishinq of prejudice and neqative cultural stereotypes in intercultural communication in South Africa. The siqnificance of this explanation is explored further by showinq how such an interactional account fits into a more comprehensive explanation of the causes of discrimination in South Africa, one that includes, also. structural explanations, and explanations in terms of the psycholoqy of individuals. This prepares the way for a consideration of the possible contribution of interactional sociolinquistics to solutions to the problem of discrimination both in South Africa and elsewhere. / Thesis (Ph.D.)-University of Natal, Durban, 1987.

This dissertation, a qualitative case study, investigates issues of language and identity among sixteen young black women studying at Technikon Natal. I examine the ways in which identities are structured by discourses of which language practices are an important part. The research participants' need to learn English is also interrogated against a scenario of South Africa during and post-apartheid. In Chapter 1, I give historical background on the educational structures and legislation which affected my research participants before 2000, and then briefly describe their present context of study. The literature on which my dissertation is based is reviewed in Chapter 2. Humanist theories of motivation for second language learning, for example, Gardner's (1985), are rejected in favour of Norton Peirce's (1995) notion of investment in second language learning, which builds on Bourdieu's (1977 - 1991) concepts of capital, and views the second language learner as inseparable from her social world. However, the emphasis in Chapter 2 is given to some of the central ideas of poststructuralist thinking, particularly those pertaining to the undermining of totalising theories, and those arguing in favour of multiple subjectivity. Chapter 3 contains both my research methods, which were mainly social constructionist, as well as the broad discourse analysis techniques I deployed for my data analysis. Notable in Chapter Four, in which I present and analyse my findings, is the power of ethnic discourses to govern the use of their own and other languages by their subjects. Significant, too, are the shifts in subjectivity which individuals experience as they integrate new discourses into their lives. Amongst my conclusions in the final chapter is the notion that, from a poststructuralist perspective, code switching may allow the simultaneous display of more than one identity; that the use of English by black South Africans is fraught with contradictions, and that indigenous African languages appear vulnerable to the pressure from powerful English discourses. I also discuss the limitations of this research and make recommendations for future research, and end with the particular insights into language and subjectivity I have gained as a lecturer in my current teaching context. / Thesis (M.A.)-University of Natal, Pietermaritzburg, 2002.

This thesis reports on an investigation of the schemata of trainee teachers from a range of different ethnic and language groups in KwaZulu-Natal who had been educated in racially segregated school systems. Informed by the insight that schemata are the products of life experience and that they constrain linguistic choices (see Tannen 1979), it was hypothesised at the outset that different ethnic groups have some different assumptions of what constitutes appropriate classroom behaviour and that this schematic knowledge is reflected in the surface linguistic forms used by teachers and pupils in classroom discourse. These differences in schemata could have unfortunate consequences for pupils of a different ethnic group from their teacher, and, in particular, those pupils from historically disempowered groups. Data was collected using an eclectic mix of quantitative and qualitative methods. Firstly, students responded to a questionnaire which elicited responses concerning pupil and teacher roles. This was followed by interviews with selected student teachers during which they were asked to comment on those statements in the questionnaire which exhibited the greatest differences between respondents who attended schools administered by racially different educational authorities. Finally, a story recall experiment was conducted. Respondents/subjects were all trainee teachers at a multi-racial college of education. The analyses of the findings of the quantitative questionnaire revealed significant differences between subjects from different education systems. The interview data, however, revealed that the differences were less marked than the findings of the questionnaire suggested. The analyses of the recall experiment suggested that while some differences between the subjects who had attended schools administered by racially segregated authorities do exist, these are not as great as initially hypothesised. Teachers need to be made aware of the problems inherent in cross-cultural encounters, and this awareness should be extended to pupils. This awareness, together with goodwill, should ensure that pupils having different schemata from their teacher and/or other pupils in the classroom will not be disadvantaged. / Thesis (M.A.)-University of Natal, 1997.

The end of apartheid in South Africa in 1994 triggered unprecedented changes in the country's institutions including the school. In the city of Durban one such change was the influx of Zulu-speaking pupils into previously "Indian" and "White" schools in their quest to learn through the medium of English only. The majority of these students are less proficient in English and therefore find it difficult to participate in classroom activities. Drawing on questionnaires, interviews and personal observations of classroom interaction, this study reports on one teaching method, peer tutoring, that some teachers at Southlands Secondary use to attend to the communicative needs of these students. In particular, the study reports on how peer-tutoring works at this school, what its benefits are to the learners, what the learners' attitudes are toward this teaching method, and what its implications are for the English-only argument. The study shows that contrary to the English-only argument, using the students' native tongue, Zulu, in an English-only classroom can assist rather than impede ESL learning. Peer tutoring not only contributes to the academic development of Zulu-speaking pupils and fosters friendships and meaningful contacts between Zulu-speaking and Indian pupils, but it also provides the latter with opportunities to learn Zulu and to appreciate the language as a resource in an English-only environment. It is hoped that this study, which is very much pilot in nature, will help highlight issues that can become the subject of more detailed studies in this field. / Thesis (M.A.)-University of Natal, 1999.

The aim of this dissertation is to contribute towards research in the area of Second Language Learning and Teaching, with particular focus on English Second Language (ESL) Learning and Teaching in the context of a South African distance college. This report investigates the notion of 'best practice' in language classrooms and compares this notion with a sample of teachers' own views about what constitutes 'best practice' in language teaching. Included in this report are critical discussions regarding language teacher education and the pervasive influence of different language policies in South Africa with regard to the way such policies have influenced teachers' beliefs about their practice. There is also a focus on the debate surrounding the prescription of a single 'best practice' in teacher education. The investigative approach used in this research was essentially qualitative and this report includes details on the benefits and challenges of the narrative task as a research tool, as well as much authentic material in the form of student responses. This investigation found that while many teachers do support practices which are in keeping with official notions of 'best practice' as described in the South African language-in-education policy and Curriculum 2005, there are significant numbers of teachers who advocate practices for language teaching which seem to contradict the notion of 'best practice' including Subtractive Bilingualism, Audiolingual methodology, rote learning and even coersion. Based on these findings, recommendations for the upgrading of existing teacher education programmes and the development of new programmes include the following: • information on changing policies and practices. • credibility in change through practice. • enactment of a process syllabus. • skills development in general classroom practice. • language development as part of teacher education. • acceptance that there is no 'best method'. • the development of broad critical reflexive practice in teachers. / Thesis (M.A.) - University of Natal, Pietermaritzburg, 2001.