M.A. (African Languages) / Please refer to full text to view abstract

Le fait que toute activité linguistique et toute interactivité soient relatives à un jeu de langage / un genre / un type de conversation a été reconnu depuis longtemps par divers spécialistes. Néanmoins, jusqu'à présent, aucune théorie systématique n'a pas été construite qui peut fournir en une manière formelle la composition et la gamme des types de conversation actuels et possibles. Dans cette thèse, nous adoptons une approche topologique pour classifier les conversations et développons une théorie formelle des types conversationnels dans le cadre de Type Theory with Records. Énoncés non phrastiques (ENP) - des énoncés fragmentaires qui sont des phrases sans prédicat mais qui expriment néanmoins un sens complet dans un contexte donné - sont un phénomène caractéristique de la conversation. Dans cette thèse, nous étudions les ENP dans un corpus chinois et examinons la distribution des ENP dans les genres parlés du Corpus National britannique (BNC). Cette thèse aborde le sujet de l'élaboration d'une théorie des types conversationnels qui peut expliquer la résolution des énoncés non phrastiques à travers des différents types conversationnels. En revanche, nous testerons l’hypothèse que la variation entre des distributions des énoncés non phrastiques peut servir à structurer l’espace des types conversationnels. / That all linguistic activity and interactivity is relative to a domain/language-game /genre/conversational type has long been acknowledged by various scholars. Nonetheless, hitherto there has been no systematic theory that tries to propose a formal account regarding the make-up and range of actual and possible conversational types. In this thesis, we take a topological approach to classifying conversations and develop a formal theory of conversational types in the framework of Type Theory with Records (TTR). Non-sentential utterances (NSUs)—fragmentary utterances which are incomplete sentences but nevertheless convey a complete sentential meaning in the given context—are a characteristic phenomenon of conversation. In this thesis, we study NSUs in a Chinese corpus and investigate the distribution of NSUs across the spoken genres of the British National Corpus (BNC).This thesis tackles the topic of developing a theory of conversational types that can be used to explicate the resolution of NSUs across different conversational types. Conversely, we investigate whether the variation in the distribution of NSUs can serve as a means of structuring the space of conversational types.

No description available.

Esta pesquisa investiga propriedades da estrutura sentencial do português brasileiro, mais precisamente aquelas que desencadeiam a inversão de constituintes para a periferia à esquerda da sentença. Sustenta-se num corpus para a obtenção de dados descritivos, constituído por dez peças de teatro, sendo cada uma delas escrita por um escritor nascido no Brasil numa década diferente, abrangendo dois séculos: XIX e XX. O objetivo desta pesquisa é identificar os tipos de constituintes que ocupam a periferia esquerda de sentenças e o porquê de ocuparem tal posição, isto é, se desempenham alguma função discursiva como tópico, foco ou se simplesmente se encontram em proeminência estrutural, a fim de entender melhor como a estrutura da informação interage com a sintaxe desse dialeto. Dessa forma, o estudo das funções discursivas desempenhadas pelo sujeito, tanto na ordem SV como na ordem VS, também precisou ser considerado. O quadro teórico adotado para sua realização sustenta-se nas implementações do programa minimalista em Chomsky (2000, 2001 e 2004) feitas à Teoria de Princípios e Parâmetros. Além disso, como proposta cartográfica, adota-se Rizzi (2002, 2004) e, como conceito de tópico e de foco, o proposto em Lambrecht (1996). / This research investigates properties of the Brazilian Portuguese sentence structure, more precisely those that trigger the inversion of constituents to the left-periphery of the sentence. It is based on a corpus as a means of acquiring descriptive data, composed of ten plays, being each of them written by a writer born in a different decade, comprising two centuries: XIX and XX. The aim of this research is to identify the types of constituents that occupy the leftperiphery of sentences and why they do so (if they fulfill any discourse function like topic, focus or if they are merely in structural prominence), in order to cast light upon the way the information structure interacts with the syntax of that dialect. For that reason, it was also necessary to investigate the discourse functions played by the subject both in SV and VS orders. The theoretical framework used for the accomplishment of that task is based on the implementations made by the Minimalist Program (Chomsky - 2000, 2001 and 2004) to the Theory of Principles and Parameters. Besides, as cartographic proposal, it adopts Rizzi (2002, 2004) and, as concept of topic and focus, those proposed in Lambrecht (1996).

Esta pesquisa investiga propriedades da estrutura sentencial do português brasileiro, mais precisamente aquelas que desencadeiam a inversão de constituintes para a periferia à esquerda da sentença. Sustenta-se num corpus para a obtenção de dados descritivos, constituído por dez peças de teatro, sendo cada uma delas escrita por um escritor nascido no Brasil numa década diferente, abrangendo dois séculos: XIX e XX. O objetivo desta pesquisa é identificar os tipos de constituintes que ocupam a periferia esquerda de sentenças e o porquê de ocuparem tal posição, isto é, se desempenham alguma função discursiva como tópico, foco ou se simplesmente se encontram em proeminência estrutural, a fim de entender melhor como a estrutura da informação interage com a sintaxe desse dialeto. Dessa forma, o estudo das funções discursivas desempenhadas pelo sujeito, tanto na ordem SV como na ordem VS, também precisou ser considerado. O quadro teórico adotado para sua realização sustenta-se nas implementações do programa minimalista em Chomsky (2000, 2001 e 2004) feitas à Teoria de Princípios e Parâmetros. Além disso, como proposta cartográfica, adota-se Rizzi (2002, 2004) e, como conceito de tópico e de foco, o proposto em Lambrecht (1996). / This research investigates properties of the Brazilian Portuguese sentence structure, more precisely those that trigger the inversion of constituents to the left-periphery of the sentence. It is based on a corpus as a means of acquiring descriptive data, composed of ten plays, being each of them written by a writer born in a different decade, comprising two centuries: XIX and XX. The aim of this research is to identify the types of constituents that occupy the leftperiphery of sentences and why they do so (if they fulfill any discourse function like topic, focus or if they are merely in structural prominence), in order to cast light upon the way the information structure interacts with the syntax of that dialect. For that reason, it was also necessary to investigate the discourse functions played by the subject both in SV and VS orders. The theoretical framework used for the accomplishment of that task is based on the implementations made by the Minimalist Program (Chomsky - 2000, 2001 and 2004) to the Theory of Principles and Parameters. Besides, as cartographic proposal, it adopts Rizzi (2002, 2004) and, as concept of topic and focus, those proposed in Lambrecht (1996).

Cette recherche a pour objectif la description outillée de la pratique actuelle de l’alternance codique dans l’interaction orale en classe bilingue au Burkina Faso et de la problématique didactique liée à cette pratique dans une perspective d’amélioration. Dans une démarche de la linguistique de corpus, les outils linguistiques CLAN et Praat sont réadaptés aux investigations en contexte bilingue sur la base des théories formelle et fonctionnelle de la linguistique. Sur trois ans, 20h 36mn de films constitués de séquences de classes et d’entretiens ont été collectées. Le corpus a pris en compte des séquences des cours de disciplines linguistiques et de disciplines non linguistiques dans toutes les classes des écoles visitées. Après une transcription standardisée avec le logiciel CLAN, des codages spécifiques à notre orientation théorique ont été ajoutés à ce logiciel. Le logiciel Praat d’analyse phonétique et acoustique a été associé pour cerner les difficultés phoniques des élèves. De façon générale, au niveau quantitatif, l’emploi de L1 et de L2 tend à suivre les orientations globales officielles du pourcentage d’utilisation du français et des langues locales. Au niveau qualitatif, les résultats font ressortir, de l’analyse de l’alternance codique en interphrastique, en intraphrastique et en extraphrastique, les fonctions linguistiques, discursives, didactiques (chez les enseignants) et acquisitionnelles (chez les élèves). Des propositions d’amélioration des pratiques de classe ont été apportées. Toutes les hypothèses, sur le discours des élèves, des enseignants et sur les logiciels, ont été mises à l’épreuve. / This research aims at describing the current practice of code-switching in oral interaction in bilingual classes of Burkina Faso and didactical problems linked to this practice in a perspective of improvement. By a process of corpus linguistics, language tools, CLAN and Praat, are adapted to our bilingual context investigations on the basis of formal and functional theories of linguistics. For three years, 20h 36mn of movies of classroom activities and interviews were collected. The corpus has considered sequences of language arts classes and non-language subjects in all school classes we visited. After a standardized transcription with the CLAN software, specific codings suited to our theoretical orientation have been added to the software. Praat, a phonetic and acoustic analysis software, was associated to analyse pupils phonic difficulties. Quantitatively, the use of L1 and L2 tends to adhere to the official guidelines of the overall percentages of use of French and local languages. Qualitatively, the results of the analysis of inter-sentential, intra-sentential and extra-sentential code-switching highlight the linguistic, discursive, and didactic (for teachers) and acquisitional (for pupils) functions. Proposals for improvements were made. All hypotheses on pupils and teachers discourse, and on the software, have been tested.

Given a sentence, a paraphrase generation system produces a sentence that says the same thing but usually in a different way. The paraphrase generation problem can be formulated in the machine translation paradigm; instead of translation of English to a foreign language, the system translates an English sentence (for example) to another English sentence. Quirk et al. (2004) demonstrated this approach to generate almost 90% acceptable paraphrases. However, most of the sentences had little variation from the original input sentence. Leveraging syntactic information, this thesis project presents an approach that successfully generated more varied paraphrase sentences than the approach of Quirk et al. while maintaining coverage of the proportion of acceptable paraphrases generated. The ParaMeTer system (Paraphrasing by MT) identifies syntactic chunks in paraphrase sentences and substitutes labels for those chunks. This enables the system to generalize movements that are more syntactically plausible, as syntactic chunks generally capture sets of words that can change order in the sentence without losing grammaticality. ParaMeTer then uses statistical phrase-based MT techniques to learn alignments for the words and chunk labels alike. The baseline system followed the same pattern as the Quirk et al. system - a statistical phrase-based MT system. Human judgments showed that the syntactic approach and baseline both achieve approximately the same ratio of fluent, acceptable paraphrase sentences per fluent sentences. These judgments also showed that the ParaMeTer system has more phrase rearrangement than the baseline system. Though the baseline has more within-phrase alteration, future modifications such as a chunk-only translation model should improve ParaMeTer's variation for phrase alteration as well.

This qual-quan case study investigated the role of code-switching (CS) in education in four senior secondary schools in Botswana. CS is a communicative strategy used in many places, including Botswana, during formal and informal social occasions. CS also occurs in education; however, its occurrence is viewed as a somewhat problematical phenomenon – that it signals the speaker’s lack of proficiency in the Language of Learning and Teaching (LoLT). The study also investigated if CS in the classroom contravenes the country’s Language-in-Education Policy (LiEP), which states that English is the medium of instruction throughout the education system (Botswana Government White Paper No.2 of 1994). The study found that CS occurrence in teaching and learning has positive and negative educational effects. However, its use has adverse implications for the LiEP of Botswana. Consequently, recommendations are made on the effective use of CS and on the revision of the LiEP. The study is divided into nine chapters. Chapter One is the introduction and covers: Botswana’s language situation, including the status of English generally and in education in particular, the statement and analysis of the problem, research questions and the importance of the study. Chapter Two gives a comprehensive review of the literature on CS generally and CS in education in particular. The key words are: code-switching, code-mixing, borrowing, nonce-borrowing and re-borrowing / double-plural. Botswana’s LiEP is also discussed with respect to language planning, education and educational development. Chapter Three discusses the research design and the data-collection methods. These include: the research sites, sample selection and sampling procedures, data-collection instruments and their administration, and the independent and dependent variables used in data-collection. The importance of pre-testing the research instruments, ethical aspects observed and problems encountered during the data-collection stage are also highlighted. The role of the University of Pretoria’s Statistics Department is also explained. Hymes’ mnemonic of SPEAKING used in the analysis of the qualitative data is also described. Chapter Four presents the quantitative analysis of the respondents’ demographic details, and highlighting the differences and similarities identified. Chapters Five and Six present the results from the quantitative analysis of the teachers’ and learners’ data. The former presents the teachers’ evaluation of the learners’ language proficiency in class; the latter presents the learners’ subjective self-evaluation of their own English proficiency and their evaluation of teachers’ proficiency in English. Furthermore, both chapters respectively present the teachers and learners’ views on the role of English, Setswana and other indigenous languages in education as LoLT, and their attitude towards CS in education. The significance or the non-significance of the analyzed results is also presented. Chapter Seven presents the results from the qualitative analysis of the data (through the application of Hymes’ mnemonic of SPEAKING) obtained through lesson observations. Chapter Eight deals with the interpretation and discussion of the results through answering the main research questions. Chapter Nine presents the study’s summary, conclusions and recommendations on CS in the classroom and on Botswana’s LiEP. The study’s limitations and implications for further research are also discussed. / Thesis (DPhil)--University of Pretoria, 2011. / Afrikaans / unrestricted

Nesta tese estudamos o movimento de sintagmas preposicionados (PPs), pronomes locativos (Locs) e pronomes locativos preposicionados (LocPs) em estruturas encaixadas do português quatrocentista, admitindo que esse período acomode variações que apontam para uma competição de gramáticas na linha de Kroch (1989, 1994, 2003). Assumimos com Martins (1994), Muidine (2000) e Moraes de Castilho (2005), que os pronomes locativos hi e ende tenham coexistido com os PPs, mas foram suplantados por estes últimos a partir do séc.XV. Tomando-se como corpus representativo do português quatrocentista a Crônica do Conde Dom Pedro de Menezes de Zurara, de fato atestamos PPs convivendo com pronomes locativos nos mesmos contextos sintáticos e inclusive os redobrando. As estruturas de redobro de um locativo por um PP mostraram-se relevantes por indicar a especialização das posições desses elementos. Fazemos uso de duas linhas teóricas, conforme proposta de Kato e Tarallo (1989): um modelo mais formal como a Teoria de Princípios e Parâmetros (Chomsky, 1986) e um de cunho variacionista como a Sociolinguística Laboviana, Para uma análise descritiva dos dados, seguimos a metodologia da Sociolingüística Quantitativa, pontuando fatores lingüísticos condicionadores da posição desses constituintes. De modo geral, os cálculos estatísticos registraram altos índices de movimento dos locativos, por um lado, mas baixos índices de movimento dos PPs, por outro. Já os LocPs equilibraram movimento e posição canônica. Outros fatores como função sintática e tipo de oração incidiram no movimento de PPs e Locs, sobretudo. A análise interpretativa dos resultados se deu sob uma perspectiva gerativista, à luz da Teoria de Princípios e Parâmetros (Chomsky, 1981,1986). Tomando-se o clítico e o sujeito como marcadores de fronteira, conforme propostas de Martins (1994) e Belletti (2002), foi possível vislumbrar quatro posições de ancoragem para os Locs e PPs, que se distribuíram ao longo da estrutura sentencial. Ficou, por fim, evidente, que o movimento de Locs, PPs e LocPs é um caso de competição de gramáticas no português quatrocentista, que acompanha algumas mudanças que desembocam na constituição do português brasileiro. / This research analyses the movement of prepositioned phrases (PPs), locative pronouns (Locs) and prepositioned locative pronouns (LocPs) in embedding structures of the 15th century Portuguese language, admitting that this period has variations which point to a grammar competition in Krochs (1989, 1994, 2003) framework. According to Martins (1994), Muidine (2000) and Moraes de Castilho (2005), latin locative pronouns hi and ende, and PPs had coexisted in Portuguese until the 15th century, when PPs supplanted the usage of locative pronouns. Indeed, taking the Dom Pedro de Menezes Cronic of Zurara (Brocardo, 1997) as the 15th century Portuguese representative corpus, we have found PPs coexisting with locatives in the same syntactic contexts, redoubling them as well. The redoubling structures have turned out to be relevant for indicating the specialization of the positions of these elements. Taking Kato and Tarallo (1989) into consideration, two frameworks are used: one formal model like the theory of Principles and Parameters, and abnother variational like Labov Sociolinguistics. The Quantitative Sociolinguistic methodology is considered for a descriptive analysis of the data, indicating linguistic facts that could affect the position of PPs and locative pronouns. Statistics have registered high percentages regarding the movement of locative pronouns on one hand, but low percentages of PPs movement on the other. LocPs registered similar numbers for movement and for no movement. Other factors such as syntactic function and kind of sentence influenced the movement of PPs and Locs.. The interpretative analysis of the results has happened in the framework of Generative Syntax, considering the Principles and Parameters Theory (Chomsky 1981, 1986). Taking clitics and subject as frontier demarcators, as stated by Martins (1994) and Belletti (2002), it has been possible to verify four positions to host Locs and PPs, all distributed along the sentential structure. In conclusion, it has been evident that PP and locative movement is a grammar competition case in the 15th century Portuguese language and it will accompany some changes which will arise in the Brazilian Portuguese language.

O objetivo deste trabalho é ampliar o conhecimento translinguístico sobre a categoria modo estudando-a em uma língua indígena brasileira. A motivação deste estudo se dá principalmente porque pouco se sabe a respeito dessa categoria em línguas não pertencentes ao tronco indo-europeu (Palmer, 2001). O objeto de estudo desta pesquisa é a língua Karitiana (família Tupi, subfamília Arikém) e o corpus analisado é formado por dados provenientes dos mitos e narrativas dessa língua e de dados coletados pelo autor da pesquisa com falantes nativos. A metodologia de coleta de dados foi a elicitação contextualizada defendida por Matthewson (2004) e Mendes (2014). Modo em Karitiana foi primeiramente analisado por Storto (2002). A autora afirma que a língua possui um sistema de modo bastante desenvolvido, porém, ainda pouco compreendido. Para ela, essa língua possui seis morfemas de modo: na(ka)-/ta(ka)- (declarativo), pyt- (assertivo), pyn- (deôntico), iri- (citativo), jy- (condicional) e a/-/-y (imperativo). Esses morfemas ocorrem entre o morfema de pessoa e a raiz verbal como observado em yn a-taka-hit-ø kat (glosa: eu 2p-dec-dar-nfut isso, tradução: eu te dei isso (Storto, 1999)). Nessa primeira análise, Storto (com. pess.) classifica esses morfemas como modo porque, segundo ela, eles marcam diferentes tipos de sentença na língua. A semântica e a pragmática formal foram utilizadas como embasamento teórico da pesquisa. Alguns trabalhos assumem que modo é um morfema que marca modalidade (Bybee, 1985; Palmer, 1986). Para a semântica formal modalidade é uma categoria do significado que está relacionada à expressão de necessidades e possibilidades (Kratzer, 1981; von Fintel, 2006; Hacquard, 2011). Outros trabalhos consideram que modo é um morfema que marca tipos de sentença. Na pragmática tipos sentenciais estão relacionados à força ilocucionária da sentença (Saeed, 2009; Portner, 2011). Seguindo a terminologia de Portner (2011), esta dissertação se refere aos morfemas de modo que estiverem relacionados à expressão de modalidade como modo verbal e os morfemas de modo que estiverem relacionados ao tipo sentencial são chamados modos sentenciais. A análise dos morfemas do Karitiana classificados como modo mostrou que essa língua possui dois lugares na estrutura morfossintática do verbo para marcar a categoria \'modo\' e não apenas um como assumido anteriormente como ilustrado por a-ta-jy-hit-ø celula-ty (glosa: 2p-dec-con-dar-nfut celular-obl tradução: eu te daria um celular). Esta pesquisa assume que cada posição marca um tipo específico de modo: os morfemas que ocorrem na primeira posição (e.g. na(ka)-/ta(ka)-) marcam tipos sentenciais, ou seja, são modos sentenciais e os morfemas que ocorrem na segunda posição (e.g. pyn- e jy-) marcam modalidade, ou seja, são modos verbais. Modos verbais e modos sentenciais podem coocorrer o que é uma evidência da existência de duas posições. O estudo da categoria modo em Karitiana possibilitou um melhor entendimento dessa categoria translinguisticamente. Segundo Sadock & Zwicky (1985) morfemas de modo não coocorrem e esta dissertação mostra que eles podem ocorrer se não estiverem ambos relacionados a força ilocucionária ou modalidade. / This research aims to increase the crosslinguistic knowledge about the categories mood within a study in a Brazilian indigenous language. The reason for this study is that there are not many reliable studies of this category in unfamiliar languages (Palmer, 2001). The object of study of this research is Karitiana language (Tupi family, Arikém subfamily) and the corpus analyzed here is composed by data from the miths and stories of this language as well as data collected by the author of the research from native speakers. We used contextualized data elicitation proposed by Matthewson (2004) and Mendes (2014). Mood in Karitiana was firstly analyzed by Storto (2002) who states that this language has a quite developed mood system that is at the same time not well understood. For her, this language has six mood morphemes: na(ka)-/ta(ka)- (declarative), pyt- (assertive), pyn- (deontic), iri- (citative), jy- (conditional) and a/-/-y (imperative). These morphemes occurs between the person morpheme and the verbal root as can be seen in yn a-taka-hit-ø kat (gloss: I 2p-dec-give-nfut that, translation: I gave you that (Storto, 1999)). In this first analysis, Storto (p.c.) assumes that those morphemes mark different types of sentence, being sentential mood morphemes. We used formal semantics and pragmatics as the theoretical background for the research. Some studies assume that mood is a morpheme which marks modality (Bybee, 1985; Palmer, 1986). In formal semantics modality is a category related to the expressions of possibilities and necessities (Kratzer, 1981; von Fintel, 2006; Hacquard, 2011). Other studies consider that mood is a morpheme which marks sentential types. In pragmatics the types of sentences are related to the illocutionary force of the sentence (Saeed, 2009; Portner, 2011). We follow the terminology used by Portner (2011) and call mood morphemes related to modality verbal mood and morphemes related to sentential type are called sentential mood. The analysis of the morphemes in Karitiana classified as mood has shown that this language has two positions in the morphosyntactical structure of the verb to the mood category and not only one as previously proposed. This can be seen in a-ta-jy-hit-ø celula-ty (gloss: 2p-dec-con-give-nfut cellphone-obl translation: I would give you a cellphone). This research proposes that each position mark a specific type of mood: Morphemes of the first position (e.g. na(ka)-/ta(ka)-) mark types of sentence, therefore, they are sentential moods and morphemes that occur in the second position (e.g. pyn- e jy-) mark modality, therefore, they are verbal moods. Verbal and sentential moods can co-occur what is an evidence for the existence of two positions. The study of mood done by this research allowed a better understanding of mood category crosslinguistically. For Sadock&Zwicky (1985), mood morphemes should not co-occur and this research has shown that they can co-occur if they are not both related to illocutionary force or modality.