Дисертації з теми "Polysemy, lexical semantics, teaching"

Thesis (Ph. D.)--University of Oregon, 2007.
Typescript. Includes vita and abstract. Includes bibliographical references (leaves 342-374). Also available for download via the World Wide Web; free to University of Oregon users.

The process of learning a foreign language can be exhausting for almost all language learners. In this study, it is aimed at providing students an accessible way to reducing commitment of errors while speaking or learning English as a second language (L2) in the most relevant and practical way, too. The Second Language Acquisition is defined as a subject which is concerned with how a language is learned and has the learner in its focus including the learner’s developing language. The main issues presented throughout the chapters were concentrated on language transfer, semantics and ambiguity. Prepositions were the most concerning analytical headline of all chapters because they are the most vulnerable part of speech students or learners of English (L2) face difficulty with. Many linguists or scholars have been working on the above scopes in order to clarify and recommend learners of English that committing errors depends on the amount of language scientific information they might have.
La adquisición de un idioma, especialmente el inglés, ha sido una pasión no solo para los estudiantes de diferentes instituciones educativas en Albania sino también para otras personas de diferentes capas sociales. Por supuesto, su adquisición no es fácil debido a los cambios estructurales léxicos y morfológicos y sintácticos que, generalmente, tienen las lenguas. Dadas las dificultades que enfrentan los estudiantes albaneses o incluso otras personas comunes en la sociedad, centré mi tesis doctoral en algunas áreas específicas de la lingüística, lo que sin duda mejorará el nivel de su adquisición, reduciendo, donde sea posible, la cantidad de errores causados por la falta de información lingüística que pueda existir en ambos idiomas Me he centrado en las áreas de interferencia (transferencia de la lengua), semántica, ambigüedad, así como en oraciones con -ing como complementos preposicionales. Cabe señalar que el estudio de la preposición es el foco principal de la tesis doctoral, ya que esta clase lingüística presenta mayores dificultades en los campos antes mencionados, debido a su frecuencia en la oración.

Thesis (MA (Afrikaans Language))--University of Stellenbosch, 1990.
199 leaves single sided printed, preliminary pages and numbered pages 1-187. Includes bibliography. Digitized using a Bizhub 250 Scanner to pdf format (OCR).
ENGLISH ABSTRACT: An evaluation of semantic information in a dictionary must take into account all aspects of meaning of lexical items. Therefore the different homonyms and polysemous senses of lexical items and the underlying semantic relations between lexical items are accounted for. A description should be given of the lexical items which form the macrostructure of HAT. These lexical items include sublexical, multilexical and free lexical items. The treatment of multilexical items in HAT is not satisfactory at present and suggestions are made for the improvement of the situation. Possible solutions for distinguishing between homonymy and polysemy are given. Suggestions to determine the possible polysemous senses of a lexical item are also made. The ordering of polysemous senses and different homonyms are discussed. It is essential to make use of a synchronic criterion to distinguish between homonymy and polysemy. The HAT's treatment of semantic relations between lexical items is not always very succesfull. The semantic relation which receives the most substantial treatment is that of semantic inclusion between lexical items. The HAT's treatment of synonyms is not without problems. More attention should be given to semantic linking and semantic opposition between lexical items. An evaluation of the system of cross-references in HAT is given, with special reference to the function of cross-references. Sometimes indistinct typographical conventions are used and they confuse the user of the dictionary. The indistinct use of typographical conventions has a detrimental influence on the transfer of information. The cross-reference system of HAT can be improved by an informative explanation in the general introduction, focusing on ways and means to use it more specifically, unambiguously and consistently. A distinction is made between various types of definitions used to define lemmas. Well known lexicographic definitions such as definitions per genus et differentia, synonymdefinitions, example definitions, circular definitions and descriptive definitions are discussed. The usage definition and two definitions which explicate a semantic plus value, the contrast and diminutive definitions, are identified. Attention is also given to the role of the general introduction in a standard synchronic dictionary. Suggestions are made to improve the general introduction of HAT.
AFRIKAANSE OPSOMMING: Wanneer semantiese inligting in 'n woordeboek ondersoek word, word alle aspekte van leksikale items se betekenis in ag geneem. Dit sluit homonimiese vorme en polisemiese onderskeidinge van leksikale items in, asook die semantiese betrekkinge wat leksikale items onderling met mekaar aangaan. Die leksikale items wat die makrostruktuur van die HAT vorm, word beskryf. Dit sluit subleksikale, multileksikale en vrye leksikale items in. Verskillende tipes multileksikale en subleksikale items word onderskei. Tans word multileksikale items in HAT nie na wense hanteer nie en voorstelle word aan die hand gedoen hoe die stelsel verbeter kan word. Moontlike oplossings vir die onderskeidingsprobleem tussen homonimie en polisemie, asook maniere waarop 'n leksikale item se polisemiese onderskeidinge vasgestel kan word, word verskaf. Aandag word ook aan die ordening van polisemiese onderskeidinge en homonimiese vorme gegee. Dit is belangrik dat 'n sinchroniese maatstaf gebruik word om tussen homonimie en polisemie te onderskei. Semantiese betrekkinge tussen leksikale items word met 'n wisselende mate van sukses deur HAT hanteer. Semantiese insluiting tussen leksikale items word die meeste beskryf. Daar bestaan egter probleme by die hantering van sinonimie tussen leksikale items. Semantiese aansluiting tussen leksikale items word min deur die woordeboek aangedui. Semantiese opposisie tussen leksikale items kan ook meer deur die HAT aangedui word. Die kruisverwysingstelsel van die HAT word ondersoek. Funksies van kruisverwysings word beskryf. Kruisverwysings word wel deur die HAT redaksie gebruik, maar hulle kan nog meer gebruik word. Soms word onduidelike tipografiese konvensies gebruik en dit het swak inligtingsoordrag tot gevolg. Die kruisverwysingstelsel van die HAT kan verbeter word deur dit in die toeligtende aantekeninge te verduidelik, spesifiek en ondubbelsinnig toe te pas en deur dit konsekwent te gebruik. Verskillende tipes definisies waarmee leksikale items gedefinieer word, word onderskei. Bekende leksikografiese definisies soos genus-differentia-definisies, sinoniemdefinisies, voorbeelddefinisies, sirkeldefinisies en deskriptiewe definisies word bespreek. Die gebruiksdefinisie en twee definsies wat semantiese pluswaardes ekspliseer, naamlik die teenstellingsdefinisie en die verkleiningsdefinisie,word geidentifiseer. Die rol van toeligtende aantekeninge in 'n handwoordeboek word bespreek. Daar word gewys op die swak toeligtende aantekeninge van die HAT. Voorstelle word gedoen oor hoe die hantering van semantiese inligting in die toeligtende aantekeninge verduidelik moet.

La tesi, en el marc de la Teoria Comunicativa de la Terminologia (TCT), es proposa observar els elements d'activació del valor terminològic en els verbs fraseològics del discurs especialitzat. Es descriuen, a diferents nivells, la sintaxi i la semàntica de verbs fraseològics d'ús comú i freqüents en textos d'economia i de genòmica, i es detecten diferències i semblances en el seu comportament en ambdós discursos d'especialitat per establir tendències.

A partir de les anàlisis, identifiquem les construccions verbals mínimes que són unitats de coneixement especialitzat (UCE) i observem si els verbs que fan de nucli sintàctic d'aquests sintagmes sofreixen canvis semàntics en els contextos especialitzats. Alhora, reinterpretem la proposta de factors d'activació d'Adelstein (2007) per als verbs, i confirmem empíricament diversos supòsits de la TCT sobre la polisèmia.

La tesi consta de tres parts: els fonaments de la recerca, l'anàlisi sintacticosemàntica del corpus, i els resultats i conclusions.
La tesis, en el marco de la Teoría Comunicativa de la Terminología (TCT), se propone observar los elementos de activación del valor terminológico en los verbos fraseológicos del discurso especializado. Se describe, a distintos niveles, la sintaxis y la semántica de los verbos fraseológicos de uso común y frecuentes en textos de economía y de genómica, y se detectan diferencias y similitudes en su comportamiento en ambos discursos de especialidad para establecer tendencias.

A partir de los análisis, identificamos las construcciones verbales mínimas que son unidades de conocimiento especializado (UCE) y observamos si los verbos que funcionan como núcleo sintáctico de estos sintagmas sufren cambios semánticos en los contextos especializados. Asimismo, reinterpretamos la propuesta de factores de activación de Adelstein (2007) para los verbos, y confirmamos empíricamente distintos supuestos de la TCT sobre la polisemia.

La tesis consta de tres partes: los fundamentos de la investigación, el análisis sintacticosemántico del corpus, i los resultados y conclusiones.
This thesis, within the frame of the Communicative Theory of Terminology (CTT), aims at observing the terminological-value activating elements of phraseological verbs in specialized discourse. Syntax and Semantics of common use and frequent phraseological verbs in Economy and Genomic texts are described at different levels, and differences and similarities in their behavior in both domain discourses are detected to establish tendencies.

From the analysis, we identify the minimal verbal constructions working as specialized knowledge units (SKU) and we observe if the verbs being the syntactic core of these groups suffer semantic changes in specialized contexts. Moreover, we reinterpret for verbs the activation factors proposal by Adelstein (2007, and we empirically confirm different assumptions of CTT on polysemy.

The thesis has three parts: the basis of the research, the syntax-semantic analysis of the corpus, and the results and conclusions

Thesis (Ph. D.)--Ohio State University, 2002.
Title from first page of PDF file. Document formatted into pages; contains xii, 320 p., also contains graphics. Includes abstract and vita. Advisor: David R. Dowty, Dept. of Linguistics. Includes bibliographical references (p. 315-320).

In Norstedts Comprehensive English-Swedish Dictionary (2000) it is said that the numbered list of senses under each headword is frequency ordered. Thus, the aim of this study is to see whether this frequency order of senses agrees with the frequencies appearing in the British National Corpus (BNC). Five English, polysemous verbs were studied. For each verb, a simple search in the corpus was carried out, displaying 50 random occurrences. Each collocate was encoded with the most compatible sense from the numbered list of senses in the dictionary. The encoded tokens were compiled and listed in frequency order. This list was compared to the dictionary's list of senses. Only two of the verbs reached agreement between the highest ranked dictionary sense and the most frequent sense in the BNC simple search. None of the verbs' dictionary orders agreed completely with the emerged frequency order of the corpus occurrences, why complementary collocational learning is advocated.

La información de clase semántica de los nombres es fundamental para una amplia variedad de tareas del procesamiento del lenguaje natural (PLN), como la traducción automática, la discriminación de referentes en tareas como la detección y el seguimiento de eventos, la búsqueda de respuestas, el reconocimiento y la clasificación de nombres de entidades, la construcción y ampliación automática de ontologías, la inferencia textual, etc. Una aproximación para resolver la construcción y el mantenimiento de los léxicos de gran cobertura que alimentan los sistemas de PNL, una tarea muy costosa y lenta, es la adquisición automática de información léxica, que consiste en la inducción de una clase semántica relacionada con una palabra en concreto a partir de datos de su distribución obtenidos de un corpus. Precisamente, por esta razón, se espera que la investigación actual sobre los métodos para la producción automática de léxicos de alta calidad, con gran cantidad de información y con anotación de clase como el trabajo que aquí presentamos, tenga un gran impacto en el rendimiento de la mayoría de las aplicaciones de PNL. En esta tesis, tratamos la adquisición automática de información léxica como un problema de clasificación. Con este propósito, adoptamos métodos de aprendizaje automático para generar un modelo que represente los datos de distribución vectorial que, basados en ejemplos conocidos, permitan hacer predicciones de otras palabras desconocidas. Las principales preguntas de investigación que planteamos en esta tesis son: (i) si los datos de corpus proporcionan suficiente información para construir representaciones de palabras de forma eficiente y que resulten en decisiones de clasificación precisas y sólidas, y (ii) si la adquisición automática puede gestionar, también, los nombres polisémicos. Para hacer frente a estos problemas, realizamos una serie de validaciones empíricas sobre nombres en inglés. Nuestros resultados confirman que la información obtenida a partir de la distribución de los datos de corpus es suficiente para adquirir automáticamente clases semánticas, como lo demuestra un valor-F global promedio de 0,80 aproximadamente utilizando varios modelos de recuento de contextos y en datos de corpus de distintos tamaños. No obstante, tanto el estado de la cuestión como los experimentos que realizamos destacaron una serie de retos para este tipo de modelos, que son reducir la escasez de datos del vector y dar cuenta de la polisemia nominal en las representaciones distribucionales de las palabras. En este contexto, los modelos de word embedding (WE) mantienen la “semántica” subyacente en las ocurrencias de un nombre en los datos de corpus asignándole un vector. Con esta elección, hemos sido capaces de superar el problema de la escasez de datos, como lo demuestra un valor-F general promedio de 0,91 para las clases semánticas de nombres de sentido único, a través de una combinación de la reducción de la dimensionalidad y de números reales. Además, las representaciones de WE obtuvieron un rendimiento superior en la gestión de las ocurrencias asimétricas de cada sentido de los nombres de tipo complejo polisémicos regulares en datos de corpus. Como resultado, hemos podido clasificar directamente esos nombres en su propia clase semántica con un valor-F global promedio de 0,85. La principal aportación de esta tesis consiste en una validación empírica de diferentes representaciones de distribución utilizadas para la clasificación semántica de nombres junto con una posterior expansión del trabajo anterior, lo que se traduce en recursos léxicos y conjuntos de datos innovadores que están disponibles de forma gratuita para su descarga y uso.
La información de clase semántica de los nombres es fundamental para una amplia variedad de tareas del procesamiento del lenguaje natural (PLN), como la traducción automática, la discriminación de referentes en tareas como la detección y el seguimiento de eventos, la búsqueda de respuestas, el reconocimiento y la clasificación de nombres de entidades, la construcción y ampliación automática de ontologías, la inferencia textual, etc. Una aproximación para resolver la construcción y el mantenimiento de los léxicos de gran cobertura que alimentan los sistemas de PNL, una tarea muy costosa y lenta, es la adquisición automática de información léxica, que consiste en la inducción de una clase semántica relacionada con una palabra en concreto a partir de datos de su distribución obtenidos de un corpus. Precisamente, por esta razón, se espera que la investigación actual sobre los métodos para la producción automática de léxicos de alta calidad, con gran cantidad de información y con anotación de clase como el trabajo que aquí presentamos, tenga un gran impacto en el rendimiento de la mayoría de las aplicaciones de PNL. En esta tesis, tratamos la adquisición automática de información léxica como un problema de clasificación. Con este propósito, adoptamos métodos de aprendizaje automático para generar un modelo que represente los datos de distribución vectorial que, basados en ejemplos conocidos, permitan hacer predicciones de otras palabras desconocidas. Las principales preguntas de investigación que planteamos en esta tesis son: (i) si los datos de corpus proporcionan suficiente información para construir representaciones de palabras de forma eficiente y que resulten en decisiones de clasificación precisas y sólidas, y (ii) si la adquisición automática puede gestionar, también, los nombres polisémicos. Para hacer frente a estos problemas, realizamos una serie de validaciones empíricas sobre nombres en inglés. Nuestros resultados confirman que la información obtenida a partir de la distribución de los datos de corpus es suficiente para adquirir automáticamente clases semánticas, como lo demuestra un valor-F global promedio de 0,80 aproximadamente utilizando varios modelos de recuento de contextos y en datos de corpus de distintos tamaños. No obstante, tanto el estado de la cuestión como los experimentos que realizamos destacaron una serie de retos para este tipo de modelos, que son reducir la escasez de datos del vector y dar cuenta de la polisemia nominal en las representaciones distribucionales de las palabras. En este contexto, los modelos de word embedding (WE) mantienen la “semántica” subyacente en las ocurrencias de un nombre en los datos de corpus asignándole un vector. Con esta elección, hemos sido capaces de superar el problema de la escasez de datos, como lo demuestra un valor-F general promedio de 0,91 para las clases semánticas de nombres de sentido único, a través de una combinación de la reducción de la dimensionalidad y de números reales. Además, las representaciones de WE obtuvieron un rendimiento superior en la gestión de las ocurrencias asimétricas de cada sentido de los nombres de tipo complejo polisémicos regulares en datos de corpus. Como resultado, hemos podido clasificar directamente esos nombres en su propia clase semántica con un valor-F global promedio de 0,85. La principal aportación de esta tesis consiste en una validación empírica de diferentes representaciones de distribución utilizadas para la clasificación semántica de nombres junto con una posterior expansión del trabajo anterior, lo que se traduce en recursos léxicos y conjuntos de datos innovadores que están disponibles de forma gratuita para su descarga y uso.
Lexical semantic class information for nouns is critical for a broad variety of Natural Language Processing (NLP) tasks including, but not limited to, machine translation, discrimination of referents in tasks such as event detection and tracking, question answering, named entity recognition and classification, automatic construction and extension of ontologies, textual inference, etc. One approach to solve the costly and time-consuming manual construction and maintenance of large-coverage lexica to feed NLP systems is the Automatic Acquisition of Lexical Information, which involves the induction of a semantic class related to a particular word from distributional data gathered within a corpus. This is precisely why current research on methods for the automatic production of high- quality information-rich class-annotated lexica, such as the work presented here, is expected to have a high impact on the performance of most NLP applications. In this thesis, we address the automatic acquisition of lexical information as a classification problem. For this reason, we adopt machine learning methods to generate a model representing vectorial distributional data which, grounded on known examples, allows for the predictions of other unknown words. The main research questions we investigate in this thesis are: (i) whether corpus data provides sufficient distributional information to build efficient word representations that result in accurate and robust classification decisions and (ii) whether automatic acquisition can handle also polysemous nouns. To tackle these problems, we conducted a number of empirical validations on English nouns. Our results confirmed that the distributional information obtained from corpus data is indeed sufficient to automatically acquire lexical semantic classes, demonstrated by an average overall F1-Score of almost 0.80 using diverse count-context models and on different sized corpus data. Nonetheless, both the State of the Art and the experiments we conducted highlighted a number of challenges of this type of model such as reducing vector sparsity and accounting for nominal polysemy in distributional word representations. In this context, Word Embeddings (WE) models maintain the “semantics” underlying the occurrences of a noun in corpus data by mapping it to a feature vector. With this choice, we were able to overcome the sparse data problem, demonstrated by an average overall F1-Score of 0.91 for single-sense lexical semantic noun classes, through a combination of reduced dimensionality and “real” numbers. In addition, the WE representations obtained a higher performance in handling the asymmetrical occurrences of each sense of regular polysemous complex-type nouns in corpus data. As a result, we were able to directly classify such nouns into their own lexical-semantic class with an average overall F1-Score of 0.85. The main contribution of this dissertation consists of an empirical validation of different distributional representations used for nominal lexical semantic classification along with a subsequent expansion of previous work, which results in novel lexical resources and data sets that have been made freely available for download and use.

Cette thèse se consacre aux enjeux de la polysémie verbale pour l’acquisition des langues secondes. Nous nous intéressons plus particulièrement à la polysémie du verbe prendre. Cette thèse poursuit deux objectifs complémentaires. D’une part, notre premier objectif est de décrire la polysémie du verbe prendre au moyen d’une analyse en sémantique lexicale dans une approche cognitive. D’autre part, notre second objectif est d’évaluer l’incidence de la polysémie du verbe prendre sur les connaissances qu’ont les apprenants de ce verbe et d’isoler les différentes acceptions de prendre mises au jour par notre analyse sémantique, qui s’avèrent problématiques pour les apprenants du français L2. Par ailleurs, nous cherchons également à savoir si les problèmes liés à la polysémie chez les anglophones peuvent être liés à l’influence translangagière.L’analyse sémantique du verbe prendre porte sur l’ensemble des acceptions prédicatives du verbe. Notre corpus d’analyse provient de données lexicales issues de plusieurs dictionnaires de référence. Afin de délimiter le corpus d’acceptions soumis à l’analyse sémantique, nous avons effectué une analyse syntaxique permettant de départager les différents emplois du verbe prendre : les constructions à verbe support, les locutions verbales et les acceptions prédicatives. Puis, nous proposons une analyse sémantique monosémique des acceptions prédicatives du verbe prendre qui s’articule autour d’un noyau de sens tripartite qui ne correspond à aucune des acceptions spécifiques du verbe, mais qui est présent dans l’ensemble de ces acceptions. Chacune des parties du noyau de sens peut faire l’objet d’un fenêtrage de l’attention, qui nous permet de mettre au jour les différents sens du verbe. Nous avons également émis des hypothèses sur les différences entre le verbe prendre et ses équivalents en anglais. Pour atteindre le second objectif de notre travail, nous avons mené une expérimentation auprès de 191 apprenants du français langue seconde. Tous les participants ont complété trois tâches expérimentales : un test de closure afin de mesurer leur niveau de compétence langagière en français, une tâche de production écrite et une tâche de jugement d’acceptabilité. Les résultats des analyses de régression logistique et multinomiale montrent non seulement que l’analyse sémantique que nous avons proposée permet de prédire la connaissance des différentes acceptions du verbe par les apprenants du français L2, mais aussi que les apprenants anglophones et allophones ont un comportement différent par rapport aux types d’acceptions du verbe prendre, comportement que nous avons pu expliquer par l’influence translangagière chez les participants anglophones
The present dissertation focuses on the acquisition of highly polysemous verbs by French as a second language learners. I am particularly interested in the polysemy of the verb prendre. This dissertation has two complementary objectives. On the one hand, the first objective is to describe the polysemy of prendre through a lexical semantic analysis within a cognitive framework. On the other hand, the second objective is to evaluate the impact of this polysemy on learners' knowledge of the verb’s different senses in order to isolate those which are the most problematic for French L2 learners. Furthermore, I also investigate whether the problems associated with polysemy for Anglophone learners can be explained through cross-linguistic influence. The semantic analysis of the verb prendre describes all the predicative uses of the verb and is based on lexical data from several dictionnaires. Prior to carrying out the semantic analysis, a battery of syntactic tests was used in order to categorize the different uses of the verb prendre into the following categories: support verb constructions, idiomatic expressions and predicative uses. I propose that the verb prendre has an abstract core meaning which can be divided into three parts. While this abstract core meaning does not correspond to any of the specific senses of the verb, it is present in all of them. I use the cognitive semantic concept of the windowing of attention in order to explain how each of the parts of the core meaning can be windowed in order to obtain the different senses of the verb. Moreover, this analysis allowed me to formulated hypotheses about the differences between the verb prendre and its equivalents in English.In order to meet the second objective of this dissertation, I conducted an experiment on 191 learners of French as a second language. All participants completed three tasks: a cloze test to measure their proficiency level in French, a guided written production task and an acceptability judgement task. The results of the logistic and multinomial regression analyses show not only that the semantic analysis proposed can predict participants’ knowledge of the different senses of the verb, but also that participants whose first language is English behave differently from those whose first language is another language, confirming the presence of cross-linguistic influence for the English-speaking participants

Les verbes de position sont décrits du point de vue d'une invariance mise au point pour chacun d'entre eux en fonction de la diversité des contextes dans laquelle on les trouve. Cela conduit d'une part à considérer que dans certains contextes leur sens n'est pas inutile au contenu informatif de l'énoncé et qu'il n'est pas équivalent à celui d'un verbe d'existence. Au contraire, l'emploi d'un verbe de position par un locuteur correspond à un choix motivé par la désignation de quelque chose de plus que l'être ou la présence du sujet. D'autre part, cette invariance mène à distinguer clairement le fonctionnement sémantique de chacun de ces verbes alors qu'ils semblent entrer en concurrence lorsqu'ils apparaissent dans des contextes visiblement similaires
Posture verbs are described from the perspective of an invariance developed for each of them according to the diversity of the contexts where they can be found. It conduces, on the one hand, to consider that in some contexts their meaning is not ancillary to the informative content of the statement, and that it is not equivalent to the meaning of a being verb. In fact, the use of a posture verb by a speaker refers to a choice motivated by something more than the being or the presence of the subject. On the other hand, this invariance facilitates clear distinction of the semantical function of each of these verbs, while they may seem to have the same meaning, when they occur in contexts which seem similar at first glance

Cette étude s'inscrit dans le champ de la sémantique lexicale et explore plus particulièrement la question de la polysémie dans le domaine nominal. Dans une première étape, on explicite les enjeux théoriques d'une telle recherche. Cela conduit à accorder une place centrale au concept d'invariant sémantique pour rendre compte de l'identité sémantique d'un nom (en langue) par-delà sa variation de sens (en contexte). Dans une deuxième étape, on circonscrit l'objet empirique – les noms de parties du corps humain en français contemporain – tout en justifiant ce terrain d'étude. Puis, on précise l'approche méthodologique. La suite de la thèse est consacrée à l'investigation empirique proprement dite. Il s'agit d'abord d'offrir une description générale du potentiel de variation sémantique des noms de parties du corps humain en français. Ensuite, c'est une analyse sémantique approfondie du nom cœur qui est proposée. D'une part, on formule un invariant sémantique susceptible de subsumer tous ses emplois (au cœur du sujet, Paul a mal au cœur, avoir à cœur de réussir...). D'autre part, on montre en quoi la diversité de ses emplois présente un caractère finalement régulier. Enfin, quatre autres noms (artère, épaule, bouche et pied) font également l'objet d'une analyse spécifique. Chacune de ces quatre études est l'occasion d'éprouver la pertinence du concept d'invariant sémantique pour rendre compte de la polysémie dans le domaine nominal
This study comes within the scope of lexical semantics. More specifically, it deals with the topic of polysemy in the nominals. As a first step, theoretical issues of such a research are clarified. It leads to focus on the concept of semantic invariant to give an insight into the semantic identity of a given noun regardless of its contextual variations. As a second step, the empirical object of this research – human body part nouns in contemporary French – is delimited. This gives an opportunity to justify the choice of these nouns as a field of research and to set out the methodological approach. The rest of the dissertation consists in the empirical investigation itself. First of all, an overall description of the semantic variation of the French human body part nouns is provided. Then, a semantic analysis of the noun cœur (“heart”) is developed: on one hand, a semantic invariant – suiting all cœur's contextual variations (au cœur du sujet, Paul a mal au cœur, avoir à cœur de réussir…) – is formulated; on the other hand, it is shown how this variation is deeply regular. Eventually, four more nouns (artère “artery”, épaule “shoulder”, bouche “mouth” and pied “foot”) are studied from a semantic point of view. Each of these four studies offers a new opportunity to test the relevance of the semantic invariant concept in order to give an account of the polysemy in the nominals

Polysemy is a formal property of a vocable, which is such that it contains more than one semantically related lexical units. Identifying the polysemic structure of vocables in a given language is a key issue in theoretical semantics and lexicography, since polysemy is at the heart of the internal structure of natural language systems. The purpose of this research is twofold. It targets the development of a contrastive approach to polysemy in order to implement a set of vocabulary learning strategies. The first chapter aims at reviewing the current state of the art in the field of polysemy. Where the term polysemy was introduced during the late XX century, the concept has its roots in the Aristotelian tradition and stimulated an open debate between the various approaches, until recent research literature. At present, there are several points of dis- agreement both among polysemy theories and lexicographical methods. The second chapter is intended to present the theoretical and methodological framework that underpins the study : the Explanatory and Combinatorial Lexicology, which is the lexical branch of the Meaning-Text Theory. Explanatory and Combinatorial Lexicology develops a lexicographical model of polysemy, based on a well-defined series of semantic and syntactic criteria. Each lexical unit of a vocable — a complex cluster of semantic, formal and combinatorial properties — is disambiguated by defining its semantic actants and its lexical relations. The set of paradigmatic and syntagmatic relations is translated in the system of lexical functions. One of the more recent developments of the Explanatory and Combinatorial Lexicology, is the French Lexical Network, an ongoing project that targets the construction of a full-scale lexical resource for the French language. It displaces the lexicographical description from a paper format to an electronic environment, through the use of the lexicographical editor Dicet. The French Lexical Network deepens the theoretical representation of polysemy, by establishing a typology of co-polysemy links. Furthermore, the French Lexical Network represents a lexicographic prototype, which can be applied to other languages. Some tween projects, developed on the French model, already exist for Russian and Spanish. Similarly, a lexicographical ressource for the Italian language, the Italian Lexical Network, is developed within the framework of this study. The third section of the study provides a polysemy-based contrastive analysis of taste vocabulary in French and Italian. As a preliminary step, a priming wordlist is drawn up of 4 704 Italian vocables. Subsequently, taste-related vocables (i.e. all the vocables which are semantically included in the lexical field of taste) are extracted and analysed. Data analysis mainly suggest two different trends : synonymous polysemes, two vocables which share the same meaning at all levels of their polysemic structure and which establish the same copolysemy links ; crossing polysemes, two vocables which intersect at some points of their polysemic structures. A contrastive polysemy approach can be exploited for pedagogical purposes. This is the aim of the last chapter, which consists in a three-session module designed for B1 students of French as a second language. Throughout a series of didactic activities, the student is expected to improve the quality of its lexical knowledge and to nip lexico-semantic mistakes in the bud.

Dissertação de Mestrado em Estudos Portugueses Multidisciplinares apresentada à Universidade Aberta
O fenómeno recente da imigração em Portugal coloca à escola portuguesa o desafio de acolher e integrar os alunos migrantes. A sua presença na escola exige uma atenção especial na aquisição/aprendizagem do Português enquanto língua não materna ou língua segunda, por ser esta a língua veicular dos conhecimentos escolares e porque do seu domínio depende a plena integração dos alunos migrantes. O presente estudo pretende conhecer as competências manifestadas, no que respeita à aquisição do léxico e, especificamente, ao domínio da polissemia, por parte de um grupo de alunos eslavos, do 3.º ciclo do ensino básico, que frequenta escolas da região de Leiria, cujas línguas maternas são o Ucraniano e/ou o Russo. Pretendemos, também, conhecer dificuldades de aquisição da língua portuguesa pelos alunos em causa, a fim de fundamentar propostas didácticas que favoreçam essa aquisição. Para a caracterização do domínio lexical apresentado pelo grupo de alunos eslavos de Português Língua Não Materna (PLNM), foram aplicados um teste de vocabulário e um conjunto de testes de conhecimento polissémico, por nós especificamente elaborados para o efeito. Os resultados alcançados pelos alunos de PLNM foram confrontados com os dos alunos de Português Língua Materna (PLM). No que respeita aos resultados dos testes linguísticos, os alunos eslavos, em média, revelam um domínio de vocabulário em Português significativamente inferior ao do grupo de referência, independentemente do seu tempo de residência em Portugal. Os resultados dos testes de conhecimento polissémico aplicados mostram que os alunos de PLNM, de um modo geral, conhecem os significados prototípicos dos itens lexicais que integram o teste, bem como os significados cujo grau de saliência se situa num nível próximo da prototipicidade dos referidos itens. Verificou-se que as dificuldades reveladas pelos sujeitos eslavos no domínio lexical se devem, por um lado, ao desconhecimento de vocábulos de uso menos comum ou de baixa frequência e, por outro lado, à falta de capacidade de expressarem com precisão os significados solicitados. A partir do nosso estudo e tendo em vista a promoção da competência lexical, designadamente no que diz respeito ao domínio da polissemia, apresentamos algumas propostas de actividades pedagógicas e didácticas susceptíveis de serem desenvolvidas no âmbito do ensino-aprendizagem do Português Língua Não Materna.
The recent immigration phenomenon in Portugal has challenged the Portuguese education system to welcome and integrate the migrant pupils in the Portuguese schools. At school, their integration deserves special attention concerning Portuguese language acquisition as the second language and the vehicular language of the school contents/syllabuses. The current research aims to recognise the linguistic competences of Slavic pupils in Portuguese lexicon acquisition, in particular, related to the polysemy domain. Our research also aims to understand difficulties of Slavic pupils in acquisition of Portuguese. This study concerns a group of Ukrainian and Russian pupils who attend schools in the area of Leiria from 7th to 9th forms. A vocabulary test and a set of tests on the knowledge of polysemy were applied to Slavic and Portuguese students, in order to characterize the Slavic students’ lexical domain of Portuguese as a second language. The results obtained in the Slavic group were compared with the results obtained in Portuguese L1 pupils. According to the statistic results of the linguistic tests, the average domain of Portuguese vocabulary by the Slavic pupils is significantly lower than that of the Portuguese students. The results of polysemy tests, based on ordinary words, show that the Slavic pupils not only know the prototypical meanings of the lexical items but also their most salient meanings. It is evident that the difficulties of the Slavic pupils in the domain of lexicon are due to two reasons: first, the unfamiliarity of the low frequency vocabulary, second, their lack of capacity to express themselves precisely. In order to promote the lexical capacity of the Slavic pupils, particularly in the domain of polysemy, we propose, according to the findings of our research, some methodological activities that can be put into practice in Portuguese language teaching as a second language.

碩士
國立清華大學
臺灣語言研究與教學研究所
107
This study aims to investigate the words expressing the ‘washing’ and ‘hitting’ concepts in Truku language through the analysis of the semantics qualia and argument structure. Futhermore, following the results of this study, we provide two teaching designs for the reference of Truku teachers. Truku makes use of three verbs to expressed the ‘washing’ action – mahu ‘washing clothes’, trima ‘washing body part’, and sminaw ‘washing appliance’. We generalize the core semantics and semantics qualia of these three words, and found that lexical semantics will affect the choice of the following noun. The words mahu ‘washing clothes’ is the Agent Focus form the root bahu, which includes the meaning of ‘beating’ and ‘washing clothes’. Following this language facts, we discuss the semantic correlation between these two meanings. Besides, the other words with the meaning ‘hitting’ in terms of their argument structure. In addition to the patient argument, sipaq, sru, and sbut can take an instrumental argument. To express ‘the strength of hitting’, we have to use bahu, beebu, tabuk in collocation with one of the verbs msru, ptuluk, forming a VV construction. If we want to convey ‘the effect of hitting’, a VV construction formed with something like dmahung ‘bruise (with bloodshot) ’, sknbrayuk ‘bruise’ collocated with the verb msru ‘hit’ is used. In the end, we apply the results of this study, incorporating traditional culture and living experience of Truku to design two teaching plans for the teaching of ‘washing’ and ‘hitting’ verbs in Truku.