In old Swahili tendi and homiletic poems about 50% of vocabulary is of Arabic origin (Bertoncini 1973), and besides single words, they include noun phrases or even whole Arabic sentences. In order to prove my point, I will discuss some verses taken from the Utendi wa Shujaka by one Hasan bin Ali from Lamu. The only extant manuscript of this epic poem in 295 stanzas was brought to Germany in 1854 by Ludwig Krapf and is kept in the Library of the Orientalistic Society in Halle. The poem is written in the Lamu dialect with many archaic features, like the incomplete palatalization of KI, the demonstratives in S- and others. But what is striking is the great amount of Arabic phrases and whole sentences, to the extent that we may perhaps speak of a case of code-switching. In fact, several verses of the poem cannot be understood properly without some knowledge of the main features of Arabic grammar, such as verb conjugation (both perfective and imperfective), verb forms (or classes), active and passive participles, noun inflection (masculine and feminine, broken plurals, construct state), personal, relative and possessive pronouns, prepositions and their combination with enclitic pronouns, numerals, conjunctions and particles, as well as word order.

Thesis (M.A.)--University of Hong Kong, 1989. / Also available in print.

The study is an in-depth examination of code-switching in the Logoli speech community in the cosmopolitan Kangemi informal settlement area on the outskirts of the city of Nairobi. The aim of the study is to investigate the sociolinguistic and structural developments that result from urban language contact settings such as Kangemi. The main objective is to identify and illustrate the social motivations that influence the tendency of the Logoli speakers to alternate codes between Lulogoli, Kiswahili and English in the course of their routine conversations as well as the structural patterns that emerge in the process of code-switching. Various methodological techniques were used in the gathering of data, including questionnaire surveys, oral interviews, tape recordings and ethnographic participant-observation techniques are highlighted. Extracts from the corpus were analysed within a theoretical framework based on two models, namely the Markedness Model and the Matrix Language Frame Model, both developed by Myers-Scotton. The study identified and interpreted, within the Markedness Model framework, the key social variables that determine code-switching behaviour among the Logoli speech community. These include age, education, status and the various social domains of interaction. In the light of these factors, the researcher was able to explain the tendency to switch codes in different settings and confirm the study’s assumption that urban-based social factors largely determine the motivations for and the patterns of code-switching. This lead to the conclusion that code-switching is not a random phenomenon but a strategy and a negotiation process that aims at maximizing benefits from interaction. Structural features of the corpus were also identified and analysed within the Matrix Language Frame Model. The assumptions of the model were tested and found to be supported by numerous examples from the data. A number of recommendations were made for further research on minority languages in Kenya and the need for language policy in Kenya to be formulated to take these language groups into consideration. / Linguistics and Modern Languages / D. Litt. et Phil. (Sociolinguistics)

Grâce à une observation ethnographique menée durant trois ans sur trois terrains d’étude différents (une agence d’architecture, une organisation intergouvernementale et un réseau social professionnel), et à partir d’une approche inspirée de l’analyse conversationnelle (Schegloff, Jefferson et Sacks, 1974)d’inspiration ethnométhodologique (Garfinkel, 1967), notre recherche vise à repérer les « méthodes » qu’utilisent les membres afin de se coordonner dans la résolution des problèmes, dans la prise de décision et dans la co-élaboration des solutions professionnelles en mobilisant pour ce faire plusieurs ressources langagières.Lorsque l’on parle de « plurilinguisme » en milieu professionnel, la question sur ses conséquences en tant que ressource ou obstacle est souvent posée. Loin de considérer la diversité linguistique comme un phénomène pré-existant à l’interaction dans des espaces professionnels internationaux, nous essayerons d’approcher cette « diversité » à partir d’une perspective analytique endogène. C’est pourquoi dans le cadre du projet européen Dynamiques des Langues et Gestion de la diversité (DYLAN), la présente thèse s’intéresse avant tout au caractère contraignant ou non du plurilinguisme pour l’accomplissement d’une activité. À cet égard, nous étudierons la manière dont les acteurs sociaux se servent des ressources langagières plurilingues pour organiser leurs actions (Mondada, 2007) et participer chacun à sa manière aux activités communicatives et traiter par là même des difficultés dues à l’utilisation d’une langue. C’est en tenant compte des limites de la dichotomie ressource ou obstacle que nous décrirons comment le plurilinguisme est une pratique ajustée aux contingences de la situation communicative d’où elle émerge.Nous nous intéressons particulièrement à l’analyse des événements de la parole où la co-existence de plusieurs « répertoires verbaux » (Gumperz, 1964) apparaît, soit comme un ensemble de ressources conversationnelles utilisées dans leur continuité linguistique, soit comme un problème résolu localement pour la progression de la communication. Dans cette thèse, nous avons identifié deux configurations principales du plurilinguisme en situation de travail. Il s’agit d’une hétérogénéité des répertoires verbaux constituée en objet du discours et exploitée par les participants dans la résolution des problèmes communicatifs. Cette hétérogénéité exhibée par les membres eux-mêmes est observable dans des séquences de thématisation du choix linguistique, dans des séquences de recherche de mots (Goodwin et Goodwin,1986) et dans des séquences de réparation (Schegloff, Jefferson et Sacks, 1977). Il s’agit aussi d’une hétérogénéité ignorant toute structuration du répertoire concerné en langues. Cela a été reconnu au sein des séquences de désaccord ou de désalignement (Stivers, 2008) et dans des séquences insérées dans des formats spécifiques de participation (Goffman, 1963, 1981) où les ressources plurilingues ne sont pas thématisées,mais bien mises à la contribution de l’accomplissement pratique de l’activité. / Based on a three-year ethnographic study in several multilingual professional settings (an architecture firm, an intergovernmental organization and a professional social network) with an approach grounded in ethnomethodologically-inspired conversational analysis (CA) studies (Garfinkel, 1967;Schegloff, Jefferson and Sacks, 1974), our research seeks to discern the methods used by members incoordinating their actions to resolve problems, making decisions, and co-elaborating professional solutions,all with the help of a variety of linguistic resources.When we speak of “multilingualism” in the professional setting, the question of its consequences as a resource or obstacle is often posed. Rather than considering linguistic diversity as a phenomenon that preexists interaction in international professional settings, we will attempt to approach this “diversity” from an endogenous analytical perspective. In the framework of the European project Language Dynamics and Management of Diversity (DYLAN), this dissertation is therefore interested above all in the constraining versus non-constraining character of multilingualism in the accomplishment of an activity. In this respect, we will study the way in which social actors use multilingual language resources to organize their actions(Mondada, 2007), participate in their own manner in communicative activities, and treat difficulties linkedwith the use of a language in this very way. By taking into account limits of the resource or obstacle dichotomy, we will describe how multilingualism is a practice adjusted to the contingencies of the communicative situation from which it emerges.We are particularly interested in analyzing speech events featuring the coexistence of several “verbalrepertoires” (Gumperz, 1964), either as a group of conversational resources used in their linguisticcontinuity, or as a problem resolved locally by the progression of communication. In this dissertation, we have identified two main configurations of multilingualism in the workplace. This entails heterogeneity inverbal repertoires represented in an object of discourse and used by participants in the resolution of communicative problems. This heterogeneity displayed by the members themselves is observable inlinguistic choice topicalization sequences, word search sequences (Goodwin and Goodwin, 1986), and repair sequences (Schegloff, Jefferson and Sacks, 1977). This also entails heterogeneity that disregards all repertoire structuring concerned in languages. This has been recognized within disagreement or disalignmentsequences (Stivers, 2008) and in sequences inserted into specific formats of participation (Goffman, 1963,1981), in which multilingual resources are not topicalized, but rather, contribute to the practical accomplishment of the activity.

The fact that people tend to alternate between languages for various communicative purposes seems to increasingly interest researchers all over the world. Thus, the linguistic phenomenon of codeswitching has been given more academic attention in recent years than ever before. This particular topic has also been infused by new research because of an ongoing pedagogic debate about whether languages other than the target language should be used in foreign languages classroom or not. The debate consists of two major opinions. On one side, adherents claim that use of non-target languages limits natural target language-input and therefore damages the learning process. On the other, it is argued that non-target languages may even be beneficial for the learning process since they carry many pedagogic opportunities with them otherwise gone lost. This paper aims to find out how, and to what extent, foreign language teachers at a Swedish for Immigrants-school codeswitch into English in class. Moreover, it is also of interest to investigate how they think about their own codeswitching and how their reasoning may reflect their codeswitching self-awareness. In search for answers to these questions, three teachers have been observed in class. Later, the teachers have been interviewed to reflect upon their own codeswitching. The study revealed clear differences in the teachers’ codeswitching and codeswitching-reasoning, However, similarities were also found, and that all three teachers shared the main objective to develop their students’ communicative competence. Moreover, they also proved to have a rather realistic picture their own codeswitching. Not only were they able to roughly estimate how, and how much, they each codeswitched. Their individual results also went in line with their reasoning to a high extent, suggesting that they all have a high degree of codeswitching self-awareness.

In old Swahili tendi and homiletic poems about 50% of vocabulary is of Arabic origin (Bertoncini 1973), and besides single words, they include noun phrases or even whole Arabic sentences. In order to prove my point, I will discuss some verses taken from the Utendi wa Shujaka by one Hasan bin Ali from Lamu. The only extant manuscript of this epic poem in 295 stanzas was brought to Germany in 1854 by Ludwig Krapf and is kept in the Library of the Orientalistic Society in Halle. The poem is written in the Lamu dialect with many archaic features, like the incomplete palatalization of KI, the demonstratives in S- and others. But what is striking is the great amount of Arabic phrases and whole sentences, to the extent that we may perhaps speak of a case of code-switching. In fact, several verses of the poem cannot be understood properly without some knowledge of the main features of Arabic grammar, such as verb conjugation (both perfective and imperfective), verb forms (or classes), active and passive participles, noun inflection (masculine and feminine, broken plurals, construct state), personal, relative and possessive pronouns, prepositions and their combination with enclitic pronouns, numerals, conjunctions and particles, as well as word order.

This dissertation is comprised of three teacher-researcher studies carried out with the intention of showing teachers how to move beyond the monolingual paradigm to build upon linguistic and cultural diversity in their everyday practice. The monolingual paradigm is linked to ideologies regarding proficiency in English as the principle means of academic success and citizenship. These studies challenge this traditional way of viewing education by treating learning "as an emerging property of whole persons' legitimate peripheral participation in communities of practice" (Lave, 1991, p. 63), whether these are interest-based communities of practice beyond the classroom or bilingual communities of practice within the classroom. In order to recognize and explain this learning and inform teaching practices, I adopt a social semiotic approach in order to explore how meaning is constructed through language, and also through social interactions with all modern aspects of society, including gesture, image, performance, and music (Kress, 2012; van Leeuwen, 2005). I explore how these interactions allow youth to create diverse identities, beyond immigrant, refugee, limited English proficient, learner, and "other", in three educational arenas: 1) Outside of the classroom in interest-based communities of practice at school, 2) in a secondary dual-language content classroom, and 3) online in an educational transnational telecollaboration project. In all three studies I triangulate quantitative data of student participation and academic achievement with qualitative participant narratives and teacher-researcher observations. What results is insight into the impact of creating multimodal trajectories for youth to perform identities and knowledge as language users in schools, where historically messages of youth's social identities are ascribed in much more constricting ways (Harklau, 2003). Viewing these youth as language users, rather than learners, sends a message to both educators and youth that in education, identity formation trumps skills development, and this can lead to higher expectations, more engaging learning, and opportunities for youth to question race-language educational legacies (Malsbary, 2014; Wenger, 1998).

The study is an in-depth examination of code-switching in the Logoli speech community in the cosmopolitan Kangemi informal settlement area on the outskirts of the city of Nairobi. The aim of the study is to investigate the sociolinguistic and structural developments that result from urban language contact settings such as Kangemi. The main objective is to identify and illustrate the social motivations that influence the tendency of the Logoli speakers to alternate codes between Lulogoli, Kiswahili and English in the course of their routine conversations as well as the structural patterns that emerge in the process of code-switching. Various methodological techniques were used in the gathering of data, including questionnaire surveys, oral interviews, tape recordings and ethnographic participant-observation techniques are highlighted. Extracts from the corpus were analysed within a theoretical framework based on two models, namely the Markedness Model and the Matrix Language Frame Model, both developed by Myers-Scotton. The study identified and interpreted, within the Markedness Model framework, the key social variables that determine code-switching behaviour among the Logoli speech community. These include age, education, status and the various social domains of interaction. In the light of these factors, the researcher was able to explain the tendency to switch codes in different settings and confirm the study’s assumption that urban-based social factors largely determine the motivations for and the patterns of code-switching. This lead to the conclusion that code-switching is not a random phenomenon but a strategy and a negotiation process that aims at maximizing benefits from interaction. Structural features of the corpus were also identified and analysed within the Matrix Language Frame Model. The assumptions of the model were tested and found to be supported by numerous examples from the data. A number of recommendations were made for further research on minority languages in Kenya and the need for language policy in Kenya to be formulated to take these language groups into consideration. / Linguistics and Modern Languages

published_or_final_version / abstract / Linguistics / Master / Master of Arts

published_or_final_version / Education / Master / Master of Education