In this dissertation, I explore the formal mechanisms underlying restrictive modification by nominals (RMN). The central claim is that RMN is dependent on how definiteness is encoded in a given language. In Greek, RMN is exemplified by extra definite determiners followed by bare adjectives, as shown in (1) below. These may precede or follow the matrix nominal. Syntactically, I argue that the determiner and the adjective may form either a restrictive or non-restrictive nominal depending on their structural position. Focusing on restrictive nominals, I argue that they are adjuncts to nP, which raise to FocP when focused. These adjuncts are small nominals, consisting of acategorial roots and n. A look at the structure of the matrix noun reveals that adjectives adjoin to NumP, as they are always prenominal. A look at genitives also suggests that Greek nouns move as high as NumP. Central to this thesis is the question of what licenses RMN. Previous analyses have correlated it with rich morphology (Lekakou and Szendrői, 2007, 2008, 2010). For them, the determiner is the spell-out of inflection, but is otherwise a semantic expletive. To these claims, I counter-argue that RMN is best viewed as being dependent on how definiteness is encoded and that the definite determiner is simply underspecified for definiteness. Assuming that definiteness consists of two components, familiarity and uniqueness, and based on data from Standard English and Scottish English, I propose that definite determiners spelling out one component, familiarity, are predicted to exhibit RMN. Familiarity and uniqueness can thus be mapped into two syntactic projections, FamP and ιP, respectively. I then propose a syntactico-semantic mechanism that derives these constructions. Hence, this research offers a modern cross-linguistic account of RMN, while it also provides us with new insights about how definiteness can be encoded cross-linguistically.

In this dissertation, I explore the formal mechanisms underlying restrictive modification by nominals (RMN). The central claim is that RMN is dependent on how definiteness is encoded in a given language. In Greek, RMN is exemplified by extra definite determiners followed by bare adjectives, as shown in (1) below. These may precede or follow the matrix nominal. Syntactically, I argue that the determiner and the adjective may form either a restrictive or non-restrictive nominal depending on their structural position. Focusing on restrictive nominals, I argue that they are adjuncts to nP, which raise to FocP when focused. These adjuncts are small nominals, consisting of acategorial roots and n. A look at the structure of the matrix noun reveals that adjectives adjoin to NumP, as they are always prenominal. A look at genitives also suggests that Greek nouns move as high as NumP. Central to this thesis is the question of what licenses RMN. Previous analyses have correlated it with rich morphology (Lekakou and Szendrői, 2007, 2008, 2010). For them, the determiner is the spell-out of inflection, but is otherwise a semantic expletive. To these claims, I counter-argue that RMN is best viewed as being dependent on how definiteness is encoded and that the definite determiner is simply underspecified for definiteness. Assuming that definiteness consists of two components, familiarity and uniqueness, and based on data from Standard English and Scottish English, I propose that definite determiners spelling out one component, familiarity, are predicted to exhibit RMN. Familiarity and uniqueness can thus be mapped into two syntactic projections, FamP and ιP, respectively. I then propose a syntactico-semantic mechanism that derives these constructions. Hence, this research offers a modern cross-linguistic account of RMN, while it also provides us with new insights about how definiteness can be encoded cross-linguistically.

Not available / text

Michif is a mixed language spoken in Manitoba, Saskatchewan, and North Dakota in which the majority of the noun phrase is derived from French and the verb phrase is derived from Cree. This thesis provides an analysis of the structure of the determiner phrase (DP) in Michif, based on data from The Michif dictionary: Turtle Mountain Chippewa Cree, by Patline Laverdure and Ida Rose Allard (1983). Even though the majority of the DP is French, Cree contributes demonstratives and quantifiers. This thesis examines the use of articles, quantifiers and discontinuous constituents(where part of the DP appears to the left of the verb and the remainder is on the right). The syntax of the Michif DP is mixed, which two syntaxes at work in which the French-derived DP is embedded within the Cree-derived DP.

Correction of the articles in English texts is approached as an article generation task, i.e. each noun phrase is assigned with a class corresponding to the definite, indefinite or zero article. Supervised machine learning methods are used to first replicate and then improve upon the best reported result in the literature known to the author. By feature engineering and a different choice of the learning method, about 34% drop in error is achieved. The resulting model is further compared to the performance of expert annotators. Although the comparison is not straightforward due to the differences in the data, the results indicate the performance of the trained model is comparable to the human-level performance when measured on the in-domain data. On the other hand, the model does not generalize well to different types of data. Using a large-scale language model to predict an article (or no article) for each word of the text has not proved successful. 1

This thesis describes the development of Dapper (`Determiner And PrePosition Error Recogniser'), a system designed to automatically acquire models of occurrence for English prepositions and determiners to allow for the detection and correction of errors in their usage, especially in the writing of non-native speakers of the language. Prepositions and determiners are focused on because they are parts of speech whose usage is particularly challenging to acquire, both for students of the language and for natural language processing tools. The work presented in this thesis proposes to address this problem by developing a system which can acquire models of correct preposition and determiner occurrence, and can use this knowledge to identify divergences from these models as errors. The contexts of these parts of speech are represented by a sophisticated feature set, incorporating a variety of semantic and syntactic elements. DAPPER is found to perform well on preposition and determiner selection tasks in correct native English text. Results on each preposition and determiner are discussed in detail to understand the possible reasons for variations in performance, and whether these are due to problems with the structure of DAPPER or to deeper linguistic reasons. An in-depth analysis of all features used is also offered, quantifying the contribution of each feature individually. This can help establish if the decision to include complex semantic and syntactic features is justified in the context of this task. Finally, the performance of DAPPER on non-native English text is assessed. The system is found to be robust when applied to text which does not contain any preposition or determiner errors. On an error correction task, results are mixed: DAPPER shows promising results on preposition selection and determiner confusion (definite vs. indefinite) errors, but is less successful in detecting errors involving missing or extraneous determiners. Several characteristics of learner writing are described, to gain a clearer understanding of what problems arise when natural language processing tools are used with this kind of text. It is concluded that the construction of contextual models is a viable approach to the task of preposition and determiner selection, despite outstanding issues pertaining to the domain of non-native writing.

This dissertation investigates nominal arguments in classifier languages (ClLs). There are two main goals. The first is to explore what is constant and what varies in the way ClLs form nominal arguments. The second goal is to understand the relationship between argument formation in classifier languages and argument formation more generally. Three classifier languages are the center of the discussion: Mandarin, a ClL without overt evidence of determiners, Yi, a head-final ClL which will be shown to have overt determiners, and Bengali, a ClL that has already been argued to have overt evidence of determiners. In addition to paying particular attention to these three ClLs, the discussion of nominal arguments also covers a wider range of ClLs and number marking languages (NMLs) from Romance, Germanic, and Slavic, as well as Hindi. In this dissertation we will argue for the following three points. First, numeral constructions (NCs) have identical syntax and semantics in ClLs and NMLs (possibly universally); specifically, we argue that NCs have a predicative interpretation and an argumental interpretation that arises via a choice function in the lexical entry of numerals. Secondly, we argue that language variation in the nominal domain is due primarily to two interrelated factors: what nouns denote (kinds or properties) and what low functional heads (i.e. number morphology (#) and classifiers) denote; we show how this variation in the nominal domain can be related to a more general macroparameter. Thirdly, we argue that determiners in ClLs are in fact expected, contrary to the standard view, but while they can combine with numeral-classifier phrases (ClPs) and numeral-less ClPs, they can never combine with bare nouns. The proposal is that bare nouns in ClLs are always argumental regardless of whether or not there are determiners. In the last chapter of this dissertation, we show that the developed analysis of nominal arguments and language variation yields an updated language typology of argument formation. With this proposed analysis of nominal arguments, we may be a few steps closer to a general theory of argument formation of wide cross-linguistic applicability. / Linguistics

The determiners, including such words as the articles, the and a, and the demonstratives, this and that, have been studied from many points of view. There are grammatical theories of their derivation and use, philosophical investigations and psychological studies, looking at adult use. However, few studies have considered the acquisition of these small, but important, words in child language development. Those studies that do exist tend to regard the child's acquisition as a progressive approximation to, or attainment of, adult usage. Chapter 1 of the thesis reviews the previous literature in the area in order to place in perspective the present research study. Chapter 1 is divided into six sections. The first section serves as a short introduction to the determiners and examines briefly some of the approaches to their study. Philosophical and linguistic studies are mentioned in passing. Historically, the definite, article the and the demonstrative that are derived from the same word in Old English (poet), while a is derived from the numeral one. However, most psychological studies of determiner acquisition have been derived from the assumption that the definite and indefinite (articles are part of one contrastive system. In order to discuss the psychological studies, section 2 examines the grammatical theories, starting with Christophersen (1939) and Jespersen (1949), of the articles and the demonstratives, as contrastive systems of language use. However, a second approach, which is taken up by developmental psycholinguists, is what is termed the functional approach. This approach is advocated primarily by psycholinguists and philosophers of language, who believe that the articles and demonstratives are linked (as they are historically) into one integrated system of determination. Section 3 therefore considers integrated theories of determiner acquisition, commencing from the work of Lyons (1975, 1977). It has been established that there are two theoretical approaches to the study of determiner acquisition, the contrastive approach and the integrated functional approach. The rest of Chapter 1 is concerned with empirical approaches, and section A examines some psychological studies of article acquisition. The work of Brown (1973), Maratsos (1976) and Warden (1973) represents studies based on the assumed contrast between the and a. The work of Bresson (1974) and Karmiloff-Smith (1976, 1979) represents the functional approach to article acquisition. The contrastive approach tends to regard the child as working towards adult competence with the articles, and thus the young child errs in his use. Brown, Maratsos, and Warden each deal at length with the child's apparent egocentric use of the definite article. The child uses the instead of a, when introducing a referent known to himself, to a listener who does not have the same knowledge. Brown draws on spontaneous speech, and considers mainly the correctness of syntactic forms. Maratsos and Warden consider the meanings of the words and the child's developing grasp of the articles as a semantic system. Bresson and Karmiloff-Smith, while both working with French-speaking children, consider the articles as part of a total system of determination. Although Bresson tends to regard the children as erring when they do not possess adult functions of the articles, Karmiloff-Smith, in a very extensive study, looks at what the children produce and understand. She then postulates the functions the determiners have for young children, how the functions are initially established and how they develop and change with an increase in linguistic and cognitive competence. The present research thesis could be viewed as an extension of this approach with English-speaking children. A similar distinction between the contrastive and functional approaches to determiner acquisition is seen with the demonstratives, and section 5 of Chapter 1 considers the work of Clark (1978). She looks at the acquisition of the demonstratives (and other deictic pairs) in terms of the child learning specific contrasts, e.g. this vs. that as proximal vs. non-proximal spatio-temporal distance. While Karmiloff-Smith also deals with the demonstratives in her functional approach, the work of Wales (1978, 1979) is discussed. Wales, while considering experimentally the acquisition of the contrastive deictic terms, also looks at spontaneous use of the determiners in mother-child interaction. Not only is the speech examined, but also the nonlinguistic gestures that accompany the speech of young children. The final section of the review chapter summarises the previous research. Also presented are the broad aims of the experiments that are reported in Chapters 2, 3 and 4. The functional approach is taken, with the. research being based on the notion that the article system and the demonstrative system are not separate and individually contrastive systems, but are linked via the and that, in their deictic functions. The experiments aimed to show how production and comprehension of the determiners can be influenced by various contextual factors. Each experiment was designed to allow for maximum flexibility, and all verbal and nonverbal responses were recorded and subjected to analysis. In this way, a clearer indication of precisely what functions of the articles and demonstratives three year old children are competent with, can be gained. Chapter 2, presenting the article comprehension experiments, commences by outlining the theoretically assumed adult functions of the articles. These functions may not necessarily be the ones on which the three year old child's article system is based. However, the functions are derived from adult-based notions of usage, so it is reasonable to suggest that they will serve as valid assumptions on which to base the experiments. Three experiments were conducted aimed at examining the young child's understanding of assumed contrasts between the functions of the and a. In all but one condition of one experiment, the children did not provide evidence of understanding the theoretically assumed contrasts. Instead, nonlinguistic response bias explanations were put forward of how the children were performing. However it is not known if the response biases arise because the children do not understand the language, or whether such biases (for absolute location, for relative location, depending on the nature of the task) block any potential understanding. Finally, the problem of designing tasks suitable for article comprehension is discussed. Chapter 3 presents the experiments designed to elicit the articles (and other determiners, both linguistic and nonlinguistic). Experiments to elicit only the articles (and their assumed contrast) tend to be fairly un-natural (see Maratsos, 1976), and hence flexible task designs were adopted for the present studies. Many forms of determiner use were elicited by these tasks, and these form the basis for Chapter 3. Experiment 4 studied the already well-documented use of the indefinite article for naming. However, a large incidence (about 30%) of article omission was recorded - an incidence which re-occurred throughout the experiments. Experiments 5 to 7 investigated the effects of various manipulations on subsequent article use. These variables included : the naming of the objects prior to subsequent questioning; altering the form of the question posed; the use of hidden vs. visible arrays; and variation of the class composition of the arrays. It was found that there was an interactive effect of these variables on subsequent article use and the functions of these article forms, but that the form of the question posed. had the greatest effect.

Notre étude s’inscrit dans le courant de réforme du système éducatif tunisien. Elle s’inscrit également dans le courant des analyses des pratiques effectives à des fins compréhensives et celui de la didactique comparée. Elle vise à analyser et comprendre les pratiques ordinaires d’enseignement de la chimie dans deux classes terminales tunisiennes lors de l’enseignement de la variation du pH au cours de la réaction entre acide et base. Egalement, elle se focalise sur les continuités et les discontinuités dans l’avancée des savoirs, entre différents registres sémiotiques, entre champs empirique et théorique, entre niveaux microscopique et macroscopique. Sur le plan théorique, nous avons combiné des descripteurs empruntés à la théorie de l’action conjointe en didactique et d’autres descripteurs spécifiques des savoirs concernés. Nous avons filmé deux enseignants l’un expérimenté, et l’autre quasi-novice dans l’enseignement à ce niveau. Les vidéos ont été retranscrites, structurées, puis analysées. Ces analyses ont permis d’établir que les enseignants observés construisent seuls les savoirs dans la classe et les exposent linéairement, seul l’enseignant expérimenté y revenant dessus pour les approfondir. Par ailleurs, l’action didactique du professeur expérimenté est caractérisée par une continuité dans l’avancée du savoir. A l’inverse, l’enseignante novice crée à son insu une multitude de discontinuités dans cette avancée. Nous avons aussi inférés certains de déterminants de l’action didactique. Ainsi, l’action des enseignants peut se comprendre en référence à l’institution « classes de terminales » dans laquelle ils enseignent. Elle peut également se comprendre au regard d’un point de vue sur l’apprentissage selon lequel les élèves apprendraient à partir d’un exposé cohérent des savoirs et d’un point de vue sur les savoirs scientifique selon lequel ces derniers ne se discuteraient pas. Ces éléments pourraient être pris en compte dans la formation des enseignants. / Our study belongs to the trend of changes in the Tunisian educational system. It also belongs to the stream of teaching practices analysis. It aims at describing and understanding ordinary chemistry teaching practices linked to the teaching of the variation of pH during the titration of the reaction acid base. It also focuses on continuity and discontinuity during knowledge progress. The study is based on the joint action theory in didactics and on descriptors related to knowledge at stake. We have filmed two teachers, one is experienced, and the other is novice. Videos have been transcribed, structured and analysed. We have shown that the observed teachers build knowledge alone, and propose it linearly. Only the experienced teacher deepens them. Moreover, the experienced teacher’s action is characterized by the knowledge continuity while the novice teacher creates in his discourse many discontinuities. Finally, their viewpoint on learning, in the nature of chemistry knowledge, the grade of the class they teach in determines their action. These elements could be used in teacher training.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-03-29T11:33:16Z No. of bitstreams: 1 sabrinaanacletoteixeira.pdf: 903908 bytes, checksum: 8a5b0ca66d22526e087950117bd1db22 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-04-24T02:45:42Z (GMT) No. of bitstreams: 1 sabrinaanacletoteixeira.pdf: 903908 bytes, checksum: 8a5b0ca66d22526e087950117bd1db22 (MD5) / Made available in DSpace on 2016-04-24T02:45:42Z (GMT). No. of bitstreams: 1 sabrinaanacletoteixeira.pdf: 903908 bytes, checksum: 8a5b0ca66d22526e087950117bd1db22 (MD5) Previous issue date: 2013-02-26 / Este estudo investiga o processo de categorização de palavras pertencentes às categorias lexicais N(ome) e V(erbo). Nosso objetivo é verificar se crianças em estágio inicial de aquisição do PB são sensíveis às categorias funcionais de determinantes e pronomes e se elas podem usar esses itens funcionais para categorizar palavras novas como nomes ou verbos. A perspectiva teórica adotada visa a conciliar um modelo de processamento voltado para a aquisição da linguagem – Bootstrapping Fonológico (MORGAN; DEMUTH, 1996; CHRISTOPHE et al., 1997) – e uma teoria linguística que considera uma interface entre o sistema linguístico e outros sistemas – sistemas sensório-motor e conceptual-intencional (CHOMSKY, 1995, 1999; HAUSER; CHOMSKY; FITCH, 2002). Tal conciliação permite-nos explicar como as crianças extraem a estrutura sintática subjacente à sua língua do continuum da fala, a partir das pistas distribucionais e prosódicas. Trabalhos anteriores com bebês alemães (HÖHLE et al., 2004) e canadenses de língua francesa (SHI; MELANÇON, 2010) obtiveram resultados que indicam que bebês de 14 meses usam os determinantes para categorizar palavras novas como nomes. Nossa hipótese é que bebês aos 13 meses identificam os determinantes e os pronomes como grupos distintos dentro da categoria dos itens funcionais e que esses itens podem guiar o processo de categorização de nomes e verbos. Os resultados de um experimento usando a técnica de Olhar Preferencial apontam para o uso dos determinantes e dos pronomes para a categorização de palavras novas como nomes e verbos, respectivamente, no PB por bebês brasileiros de 13 meses. / This study investigates the categorization process of words that belong to lexical categories N(oun) or V(erb). Our goal is to verify if children in the early stages of the acquisition of BP are sensitive to function categories such as determiners and pronouns and if they can use these function items to categorize novel words as nouns or verbs. The theoretical approach adopted aims at conciliating a processing model of language acquisition – Phonological Bootstrapping (MORGAN and DEMUTH, 1996; CHRISTOPHE et al., 1997) with a linguistic theory that considers an interface between linguistic system and other systems – sensory-motor and conceptual-intentional systems (CHOMSKY, 1995, 1999; HAUSER; CHOMSKY and FITCH, 2002). This conciliation allows us to explain how children extract the syntactic structure of their language from the speech through distributional and prosodic cues. Previous studies with German (HÖHLE et al., 2004) and French Canadian (SHI and MELANÇON, 2010) infants indicate that fourteen-month-old infants use determiners to categorize novel words as nouns. Our hypothesis is that thirteen-month-old infants identify determiners and pronouns as belonging to distinct groups within function categories and that these items can guide the categorization process of nouns and verbs. Results of an experiment using the Preferential Looking Paradigm suggest the use of determiners and pronouns to categorize novel words either as nouns or verbs in BP by thirteen-month-old Brazilian infants.