This thesis documents the use of Natural Language Processing (NLP) in Computer Assisted Language Learning (CALL) and its contribution to the learning experience of students studying Arabic as a foreign language. The goal of this project is to build an Intelligent Computer Assisted Language Learning (ICALL) system that provides computational assistance to learners of Arabic by teaching grammar, producing homework and issuing students with immediate feedback. To produce this system we use the Parasite system, which produces morphological, syntactic and semantic analysis of textual input, and extend it to provide error detection and diagnosis. The methodology we adopt involves relaxing constraints on unification so that correct information contained in a badly formed sentence may still be used to obtain a coherent overall analysis. We look at a range of errors, drawn from experience with learners at various levels, covering word internal problems (addition of inappropriate affixes, failure to apply morphotactic rules properly) and problems with relations between words (local constraints on features, and word order problems). As feedback is an important factor in learning, we look into different types of feedback that can be used to evaluate which is the most appropriate for the aim of our system.

No description available.

Diese Dissertation untersucht die lexikalischen und morphosyntaktischen Eigenschaften polnischer und russischer Prädikatsnomina sowie der Kopula (des Kopulalexems). Sie stellt die Kasusvariation (Nominativ vs. Instrumental) sowie die Formvariation (Kurz- vs. Langformadjektiv) an diesen Prädikativen in den Mittelpunkt und hat das Ziel, ihre grammatische Quelle, Funktionsweise und Interpretation bzw. Lesart zu erklären. Dabei werden sowohl die primäre als auch die sekundäre Prädikation betrachtet. Im Rahmen der Formulierung der theoretischen Grundlagen wird ein Vorschlag gemacht, der es gestattet, den viel diskutierten Begriff "Prädikation" referenzsemantisch zu erfassen. Ferner wird vorgeschlagen, Flexionsmorphologie und Semantik (im Sinne von "grammatischer Bedeutung") voneinander zu trennen. In diesem Sinne leistet die vorgelegte Dissertation auch einen Beitrag zur Forschung über die Schnittstelle zwischen Morphosyntax und Semantik.

Le créole afro-portugais de Casamance reste encore méconnu. La présente étude constitue la première description des pratiques langagières des créolophones casamançais. Elle est fondée sur un corpus de première main recueilli lors de discussions spontanées entre des locuteurs créolophones dans des villes multilingues : Dakar, Thiès et Ziguinchor. Dans ce corpus, les principales langues en contact sont essentiellement le créole casamançais, le français et le wolof. Cette thèse décrit les changements induits par ce contact dans le corpus, notamment le repérage temporel et les phénomènes touchant spécifiquement les groupes nominaux (déterminants et génitifs). L’étude de ces champs m’a permis de constater que des éléments grammaticaux et lexicaux sont fournis à la fois par l’ensemble de ces langues. Ce cas de figure n'est pas souvent pris en compte dans les approches théoriques du contact de langues, lesquelles proposent souvent une séparation fonctionnelle des langues fondée sur une dichotomie entre langue matrice et langue insérée, et tendent à ignorer les situations de contact impliquant plus de deux langues. Cette thèse constitue une contribution à l’étude du contact de langues et permet en particulier de porter un regard neuf sur une situation de contact trilingue, impliquant une langue créole et deux autres langues qui en sont typologiquement éloignées. La prise en compte d’une telle configuration revêt un caractère particulièrement novateur dans le domaine des études créoles, où les chercheurs intéressés par le contact de langues se concentrent surtout sur des situations de contact entre les créoles et leurs langues lexificatrices respectives. / Casamancese Afro-Portuguese Creole is still largely unknown. The present study is the first description of the language practices of Casamance Creole speakers. The study is based on a first-hand corpus collected during spontaneous discussions among Creole speakers in multilingual cities: Dakar, Thies and Ziguinchor. In this corpus, the main languages in contact are the Casamancese Creole, French and Wolof.This dissertation describes the changes induced by this contact in the corpus, especially the temporal deixis and phenomena specifically affecting noun groups (such as determiners and genitives). The study of these domains has led me to realize that all three languages involved both grammatical and lexical elements are provided by all these languages.Such a scenario is generally not taken into consideration in the theoretical approaches to language contact, which often posit a functional separation of the languages involved based on a dichotomy between matrix language and embedded language, and tend to ignore contact situations involving more than two languages.This dissertation is therefore a contribution to the study of language contact; more specifically, it allows for the possibility to take a fresh look at a trilingual contact situation involving a Creole language and two other languages typologically distant from the former. The inclusion of such a configuration is particularly innovative in the domain of Creole studies, where researchers interested in contact languages focus mainly on situations of contact between Creoles and their respective lexifier languages

Cette thèse aborde la reformulation dans des discussions entre jeunes Libanais francophones. Après une approche lexicologique de formuler/reformuler/reformulation qui vise à faire émerger leurs spécificités par rapport à dire/redire, énoncer/réénoncer, etc., et une tentative de trouver une unité à l’objet discursif qu’est la reformulation, une double approche des reformulations du corpus est engagée. La première, linguistique, dresse une typologie de ce phénomène dans le corpus. Se voulant globale, elle prend en compte des critères linguistiques (phonologique, sémantique, lexical, morphosyntaxiques) et d’autres qui relèvent de la dimension interactive dans laquelle s’inscrivent les reformulations (instances, initiation et rôle de la reformulation dans le discours). Le principal apport de cette approche qualitative et quantitative est d’avoir montré l’aspect central de la morphosyntaxe dans le marquage des reformulations nuançant ainsi les approches du phénomène à travers les marqueurs lexicaux et les connecteurs. La deuxième approche se centre sur les locuteurs et sur leur statut de non-natifs en s’interrogeant sur le potentiel des reformulations à nous renseigner sur leur compétence de communication. De plus, le travail se penche sur des phénomènes de contacts de langues dans le cadre de l’activité reformulatoire des locuteurs dans le corpus. Cette thèse se consacre enfin, dans le cadre théorique de la grammaticalisation/pragmaticalisation, à l’étude d’un marqueur discursif du dialecte libanais «jæʕne» qui fait irruption dans le discours des locuteurs et témoigne, entre autres emplois, de celui de marqueur de reformulation. Les emplois de jæʕne sont considérés aussi comme des phénomènes de contacts de langue d’une part dans le cadre d’alternance codique et d’autre part, à travers des calques opérés par les locuteurs sur c’est-à-dire. / This PhD thesis approaches reformulation in conversations between young francophone Lebanese. It starts with a lexicological approach of the words formulate/reformulate/reformulation in order to emphasize their differences with say/say again, utter/utter again and goes on with a deep analysis of reformulation as a discursive object, followed by a two-fold approach of the occurrences in the corpus. The first approach is a linguistic one and also deals with the interactional dimension of reformulations. Quantitative and qualitative analyses emphasize the role of morphosyntax in marking reformulations, completing the approaches based on discourse markers and connectives. The second approach, focusing on the speakers, takes into account their non-native speaker status. We are questioning here the way in which reformulations can provide information about their communicative competence. Reformulation is seen as a communicative strategy they use in order to solve communication problems. Our work ends with the analysis of a Lebanese French discourse marker, jæʕne, used by the speakers to indicate reformulation (among other things). The first part of the analysis draws a typology of all its uses within the grammaticalization/pragmaticalization framework while the second one studies it as a language contact phenomenon.

Depuis la loi du 11-02-2005 pour l’égalité des droits et des chances, les lieux ouverts au public doivent accueillir les Sourds en Langue des Signes Française (LSF). C’est dans le cadre du développement d’outils technologiques de diffusion de LSF que nous avons travaillé, plus particulièrement au niveau de la traduction automatique du français écrit vers la LSF. Notre thèse commence par un état de l’art relatif aux connaissances sur la LSF (ressources disponibles et supports d’expression de la LSF) avant d’approfondir des notions de grammaire descriptive. Notre hypothèse de travail est la suivant : la LSF est une langue et, à ce titre, la traduction automatique lui est applicable.Nous décrivons ensuite les spécifications linguistiques pour le traitement automatique, en fonction des observations mises en avant dans l’état de l’art et des propositions de nos informateurs. Nous détaillons notre méthodologie et présentons l’avancée de nos travaux autour de la formalisation des données linguistiques à partir des spécificités de la LSF dont certaines (model verbal, modification adjectivale et adverbiale, organisation des substantifs, problématiques de l’accord) ont nécessité un traitement plus approfondi. Nous présentons le cadre applicatif dans lequel nous avons travaillé : les systèmes de traduction automatique et d’animation de personnage virtuel de France Telecom R&D. Puis, après un rapide état de l’art sur les technologies avatar nous décrivons nos modalités de contrôle du moteur de synthèse de geste grâce au format d’échange mis au point. Enfin, nous terminons par nos évaluations et perspectives de recherche et de développements qui pourront suivre cette Thèse.Notre approche a donné ses premiers résultats puisque nous avons atteint notre objectif de faire fonctionner la chaîne complète de traduction : de la saisie d'un énoncé en français jusqu'à la réalisation de l'énoncé correspondant en LSF par un personnage de synthèse. / Since the law was voted the 11-02-2005 for equal rights and opportunities: places open to anyone (public places, shops, internet, etc.) should welcome the Deaf in French Sign Language (FSL). We have worked on the development of technological tools to promote LSF, especially in machine translation from written French to FSL.Our thesis begins with a presentation of knowledge on FSL (theoretical resources and ways to edit FSL) and follows by further concepts of descriptive grammar. Our working hypothesis is: FSL is a language and, therefore, machine translation is relevant.We describe the language specifications for automatic processing, based on scientific knowledge and proposals of our native FSL speaker informants. We also expose our methodology, and do present the advancement of our work in the formalization of linguistic data based on the specificities of FSL which certain (verbs scheme, adjective and adverb modification, organization of nouns, agreement patterns) require further analysis.We do present the application framework in which we worked on: the machine translation system and virtual characters animation system of France Telecom R&D.After a short avatar technology presentation, we explain our control modalities of the gesture synthesis engine through the exchange format that we developed.Finally, we conclude with an evaluation, researches and developments perspectives that could follow this thesis.Our approach has produced its first results since we have achieved our goal of running the full translation chain: from the input of a sentence in French to the realization of the corresponding sentence in FSL with a synthetic character.

Cette recherche pseudo-longitudinale privilégiant une approche sociolinguistique et didactique s’inscrit dans le domaine du développement de l’interlangue à travers l’analyse des erreurs morphosyntaxiques du français. Notre public s’est compose des lycéens kenyans anglophones. Cela a été effectué au sein de la capitale du Kenya, Nairobi et de sa banlieue. Les entretiens sur le terrain comprenant des exercices descriptifs et conversationnels à l’oral et à l’écrit, ont mené à la description, l’analyse et l’étiologie des erreurs morphosyntaxiques en français langue étrangère (désormais FLE) dans les productions des apprenants dans un contexte lycéen. L’analyse de ce corpus écrit et oral a révélé les erreurs morphosyntaxiques écrites et orales en français les plus récurrentes ainsi que leurs causes chez les lycéens kenyans. Ce travail s’achève par l’élaboration de propositions didactiques à l’écrit et à l’oral présentant des pistes de réflexion sur la remédiation des erreurs morphosyntaxiques des lycéens kenyans. / This research favoring a sociolinguistic and didactic approach is based on the development of interlanguage through the analysis of morphosyntactical errors in French by English speaking Kenyan high school students in Nairobi and its suburbs. Field research composed of descriptive and conversational oral and written exercises in French as a foreign language was carried out on selected high school students. This paved way to the description and analysis of morphosyntactical errors in French. The analysis of the written and oral corpus helped to identify the most recurrent morphosyntactical and written errors in French, as well as their etiology. This pseudo-longitudinal research culminated in suggested written and oral didactical activities aimed at the correction of these morphosyntactical errors.

Esta tese estuda quatro construções que tiveram/estão tendo mudanças em sua sintaxe no português brasileiro (PB). São elas: anticausativas, como em (1) O prato (se) quebrou: médias, como em (2) Essa roupa (se) lava fácil, o membro inacusativo da alternância agentiva, (3) Essa roupa lavou (*se) ontem e impessoais, como (4) Nesse lugar (se) vende bota. Afirmo que as mudanças nas estruturas desses eventos possuem uma origem em comum: a perda de SE que nucleia um tipo de projeção Voice específica. Essa projeção de Voice pode ou não projetar um especificador em que um argumento implícito pro é alocado. Quando essa projeção de Voice tem o especificador projetado, origina-se a interpretação de que um humano é responsável pelo evento. Essa é a interpretação em médias e impessoais com SE. Nos casos em que a projeção de Voice não tem o especificador projetado, a leitura que se tem é de um evento espontâneo que aconteceu sem que um Agente o promovesse, como em anticausativas. Em relação a esse argumento implícito em alguns dos eventos tratados, no Capítulo 2, demonstro que uma série de diferenças dos eventos estudados em relação a outras línguas românicas está relacionada ao tipo de argumento implícito projetado em Voice. O argumento implícito em PB é forte, comportando-se como um DP e funcionando como um interventor em função da sua constituição. Em contraste, o argumento implícito em espanhol, por exemplo, é fraco e não é um interventor em relações de Agree de uma sonda com mais de um argumento (nos casos em que essas relações são lícitas). Essa característica é responsável por diferenças significativas entre os eventos com SE do PB e de outras línguas românicas, algumas das quais foram exploradas nesta tese. Um caminho analítico empregado em vários dos capítulos desta tese é cotejo das variantes marcadas e não marcadas dos eventos estudados acima. No Capítulo 3, lido com anticausativas marcadas e suas contrapartes não marcadas e demonstro que nada mudou na estrutura desse evento com a perda de SE. Isso é esperado, já que anticausativas só possuem um argumento e SE tem o único papel de nuclear Voice. No Capítulo 4, estudo médias marcadas e não-marcadas e demonstro que houve uma alteração substancial na sintaxe desses eventos. Médias não-marcadas não possuem nenhum argumento implícito e são inacusativas para todos os testes relevantes. Isso mostra que as médias não-marcadas em PB se distanciam de médias não-marcadas do inglês, por exemplo, que são descritas como sentenças transitivas. Nesse caso, a perda de SE levou à perda de Voice. No Capítulo 5, estudo o membro inacusativo da alternância agentiva, exemplo em (4), e mostro que, apesar de esse tipo de evento não se combinar com SE, o que aponta que não há Voice nucleada por SE nas sentenças que expressam esse tipo de evento, elas são um subproduto da existência de médias inacusativas não-marcadas no PB. Isto é, sentenças como Essa roupa lavou ontem são uma generalização dos contextos em que uma estrutura como Essa roupa lava fácil pode entrar. Nesse sentido, essas sentenças também estão relacionadas com a perda de um tipo de Voice específico no PB, mas de uma forma indireta. Por último, no Capítulo 6, estudo sentenças impessoais e mostro que a variante não-marcada é talvez um dos eventos estudados nesta tese em que mais houve mudanças, porque essas sentenças sinalizam também uma mudança em TP. Ao contrário de médias e o membro inacusativo da alternância agentiva, todavia, sentenças impessoais necessitam de um argumento externo. Em sentenças como Nesse lugar (se) vende bota, o locativo é o especificador de Voice. Esse elemento é usado nesse tipo de sentença locativa justamente porque locativos geram leituras existenciais. Em suma, este estudo descreve uma série de mudanças em eventos do PB que acontecem, em última análise, em virtude da perda de morfologia nessa língua. O estudo contribui, assim, para dois debates maiores: i) de que forma mudanças de parâmetro podem afetar os eventos da gramática; ii) se eventos são melhor descritos por abordagens lexicais ou sintáticas. Os resultados aqui reportados mostram que mudanças de parâmetros podem afetar eventos, embora não essa não seja uma característica definidora do parâmetro pro-drop, por exemplo. Além disso, os dados do PB mostram que eventos concebidos como fruto de alternâncias lexicais em muitas abordagens atuais, como anticausativas e médias, mudam de acordo com tendências sintáticas da língua e são melhor tratadas em abordagens sintáticas em virtude disso. / In this thesis, I study four constructions that have changed in Brazilian Portuguese (BP), namely anticausatives, as in (1) The plate (se) broke; middles, as in (2) These clothes (se) wash easily; the unaccusative member of the agentive alternation, (3) This cloth (*se) washed yesterday and impersonals, as in (4) In this place (se) sell boots. I argue that all these changes are related to the loss of a Voice projection whose nucleus is the clitic SE. The said projection can or cannot project a specifier, in which pro is merged. When pro is present, the interpretation of an entity responsible for the event originates. This is found in middles and impersonal sentences with SE. As anticausatives are spontaneous events, in this particular case, this type of Voice does not project a specifier. In respect to the implicit argument in some of these events, in Chapter 2, I show that several contrasts among these events in Romance languages can be accounted for if there are different types of implicit arguments. Whereas the implicit argument in BP is strong, i.e. it behaves as a regular DP and intervenes in Agree relations, in Spanish it is a weak implicit argument and does not intervene when the probe enters in a relationship with more than one goal. These constituency differences are responsible for the differences between events with SE in BP and other languages, some of them are treated in this thesis. Throughout the analysis, I compare the SE-variant with its alternative realization without the clitic. In Chapter 3, I deal with marked anticausatives and their unmarked counterparts. I claim that nothing really changed in this case despite the loss of SE. This is expected, since Voiced anticausatives do not project an implicit argument. In Chapter 4, I deal with marked and unmarked middles, showing a substantial change in their structure. There is no implicit argument in unmarked middles and they behave as generic unaccusative for all relevant tests. This shows that unmarked middles in BP differ from English middles, which are normally analyzed as transitive or unergatives. In this case, the loss of SE led to the loss of Voice. In Chapter 5, I study the unaccusative member of the agentive alternation, exemplified by (3). I argue that, although SE is not licensed in this type of event, which shows that there is no Voice in it, they are a byproduct of unmarked middles in BP. In other words, sentences like This cloth washed yesterday (lit.) are a generalization over the contexts in which This cloth washes easily is licensed. For this reason, these sentences are also related to the loss of a specific type of Voice in BP, albeit in an indirect way. Finally, in Chapter 6, I study impersonal sentences. The unmarked variant of this type of event is perhaps the most affected one, since the changes do not affect only Voice, but also T. Unlike middles and the unaccusative member of the agentive alternation, however, impersonal sentences need an external argument. In sentences such as (4), the locative is the specifier of Voice. This element is used because locatives generate existential readings. To sum up, this study describes a series of changes in BP events that happened ultimately due to the loss of morphology in this language. This study contributes to two major debates: i) in which way changes of parameter can affect events; ii) if events are better described by lexical or syntactic approaches. The results reported in this thesis indicate that parametrical changes can affect events, even though this is not a defining property of the pro-drop parameter. In respect to (ii), BP data shows that events traditionally treated by lexical approaches, as anticausatives and middles, are tied in with general morphossyntactic changes in the language and are thus better explained by syntactic approaches.

Cette thèse comporte deux volets complémentaires : d'une part, une description phonologique et morphologique du rgyalrong zbu, langue du groupe rgyalronguique du sino-tibétain ; d'autre part, une reconstruction de l'histoire de la flexion verbale de cette langue. Le volet descriptif vise à l'exhaustivité dans la description de la phonologie de cette langue, et pose les bases d'une grammaire de référence qui ambitionne notamment de situer sa morphologie flexionnelle dans une perspective de linguistique générale. Sur la base de cette description, élaborée au fil d'enquêtes de première main sur cette langue fortement en danger, un volet diachronique est proposé. Celui-ci contribue au projet général d'une reconstruction du groupe rgyalronguique, entreprise collective qui a des implications importantes pour la reconstruction du sino-tibétain dans son ensemble. Parmi les principaux résultats figurent une reconstruction du système de marquage de temps-aspect-modalité (TAM) de l'ancêtre commun le plus récent au japhug, au tshobdun et au zbu, désigné ici comme « proto-rgyalrong supérieur ». Cette reconstruction ouvre une perspective nouvelle pour bien distinguer entre éléments hérités et développements secondaires dans chacune des langues rgyalronguiques, ainsi qu'au sein du groupement supérieur (qianguique). Le volet diachronique comporte aussi un traitement comparatif de quelques verbes du zbu, à la lumière des données des langues qianguiques et lolo-birmanes, qui pourra servir de modèle pour un dictionnaire étymologique des verbes rgyalronguiques. / This thesis focuses on Zbu Rgyalrong, a Sino-Tibetan language of the Rgyalrongic branch, and consists of both a phonological and morphological description of the language and a reconstruction of the history of its verb inflection. This thesis aims at descriptive exhaustivity for its phonology and attempts to lay the foundation of a reference grammar, in order to characterize its inflectional morphology in the perspective of general linguistics. Based on the description of this highly endangered language, the thesis contains a diachronic discussion, which contributes to the project of reconstructing Proto-Rgyalrongic, a collective enterprise which has important implications for the reconstruction of Sino-Tibetan as a whole. Notably, a reconstruction is proposed of the time-aspect-modality (TAM) marking system of Proto-Upper-Rgyalrong, the most recent common ancestor of Japhug, Tshobdun and Zbu. This reconstruction provides new perspectives for distinguishing between inherited elements and secondary developments in each Rgyalronguic language, as well as within the kindred Qiangic languages. Also included in the diachronic part is a comparative treatment of some verbs in Zbu, examined with data from other Qiangic and Lolo-Burman languages, which can serve as a model for an etymological dictionary of Rgyalrongic verbs.

La présente étude s’inscrit dans le cadre plus vaste de la description des langues de l’aire linguistique Amdo. Cette région est caractérisée par la présence de langues sinitiques, mongoliques, tibétiques et turciques et, pour le salar et le tibétain, une situation de contact linguistique long d’environ sept siècles. Le salar est l’une des langues turciques les moins décrites et elle présente de nombreuses particularités dues à son isolement par rapport aux autres langues turciques. Il n’existe pas non plus de description des variétés de tibétain parlées dans la région salarophone, périphérique dans la tibétosphère. La perspective que nous adoptons ici est donc à la fois descriptive et comparative. Après un exposé des caractéristiques historiques et sociolinguistiques de cette situation de contact, nous analysons de façon détaillée des catégories grammaticales indexées dans le syntagme nominal et dans le prédicat. En particulier, nous proposons une nouvelle analyse des morphèmes de Temps-Aspect-Mode en salar et montrons que cette langue a copié en partie ses catégories évidentielles sur le modèle de celles du tibétain de l’Amdo. Nous nous intéressons ensuite aux problématiques liées à la valence verbale et aux effets du contact linguistique sur l’organisation accusative et ergative qui caractérisent respectivement le salar et le tibétain de l’Amdo. Nous analysons les marques casuelles à la fois comme relateurs, au sein du prédicat verbal, mais également comme converbe ou au sein des formes converbiales. Enfin, nous décrivons les catégories de voix grammaticalisées en salar et en tibétain, et montrons que celles-ci sont quasiment identiques dans les deux langues. / This study falls within the larger description of the languages of the Amdo linguistic area. This area is characterized by the coexistence of Sinitic, Mongolic, Tibetic and Turkic languages, and, regarding Salar and Tibetan, an approximately seven-century-long contact situation. Salar language remains one of the less described Turkic languages and, because of its isolation from the other Turkic languages, displays many specificities. There exists no description of the Amdo-Tibetan variety spoken in the Salar-speaking region either, this region being very peripheral in the Tibetosphere. The perspective taken in this study is thus both descriptive and comparative. After a depiction of the historical and sociolinguistic characteristics of this contact-situation, we analyse in detail the grammatical categories indexed in the nominal phrase and in the predicate. Notably, we suggest a new analysis of the Tense-Aspect-Mood morphemes in Salar and we show that the Amdo-Tibetan evidential categories have been partly copied in Salar. Then, the question of verb valency is addressed, and the effects of language contact on the Turkic accusative and on the Tibetic ergative organisation are explored. We analyze the case markers not only as markers of syntactic dependancy within the verb predicate, but also in their role as or in converbs. Finally, we describe the grammatical voices attested in Salar and in Amdo Tibetan, and show that the syntactic and semantic characteristics are almost identical in the two languages.