The logic of natural language negation (e.g. no) is prima facie at odds with an exclusively CLASSICAL LOGIC based on SET THEORY. The negation of classical logic obeys the LAW OF NON-CONTRADICTION (according to which an element cannot be both in and not in a set) and the LAW OF EXCLUDED MIDDLE (according to which an element either is or is not in a set). In contrast, natural language negation tolerates APPARENT CONTRADICTIONS (as in, Do you like your supper?---Yes and no) and UNEXCLUDED MIDDLES (as in, Do you love your supper?--- No; Do you hate your supper?---No). These prima facie non-classical phenomena relate to non-set theoretic INTERPRETIVE STRUCTURES, particularly OBJECT STRUCTURE, i.e. that objects may possess properties in part but not in whole (e.g. one might love one part of a supper but hate another), and PREDICATE DIMENSIONALITY , i.e. that properties may have mid-range values (e.g. one might evaluate the supper as a whole as an average between the two extremes). I describe several experiments exploring adults' and children's grasp of object structure and predicate dimensionality and their relation to natural language negation, using two different reasoning tasks: the ability to resolve apparent contradictions and the ability to draw inferences about unexcluded middles. Results suggest that even the negation of children as young as age three is not exclusively classical and set theoretic, but rather reflects a principled grasp of object structure and predicate dimensionality. I argue that children's grasp of natural language negation is guided by sophisticated LOGICAL RESOURCES that relate to the NATURAL LOGIC of common objects and their properties, particularly object structure and predicate dimensionality.

This thesis is concerned with the syntactic properties of cleft constructions in Japanese within the Principles and Parameters framework with some consideration of the Minimalist Program. The constructions in question are divided syntactically into two sub-parts, a focus element and a presuppositional clause. This thesis claims that the focused element's position is not derived via a process of movement such as scrambling, but is instead base-generated in a predicate position occupied by a single constituent. Presuppositional clauses of clefts are argued to involve movement of a null operator, which is analyzed on a par with topicalization. This is supported by various syntactic parallelisms between cleft and topic constructions in this language. While the presuppositional clause is marked by a nominalizer, it is shown that it does not project a nominal category; in this sense, cleft constructions are analogous to head-internal relative clauses. Furthermore, one of the most interesting properties of the cleft construction, the syntactic phenomenon of 'connectivity', is closely examined. I propose that long-distance cleft constructions in Japanese have peculiar structures: a null operator originates adjoined to the highest complement clause, and its thematic position is occupied by pro. This analysis is supported by empirical facts which involve binding relations, weak crossover effects, interactions with another A'-dependency, and clefting adjunct PPs. It is shown that these types of resumptive A '-dependencies are observed across languages as well as in different constructions within the Japanese language. Finally, the discussion is extended to so-called sluicing in Japanese. This thesis observes similar syntactic behavior in sluicing and clefts, which leads to the argument that sluicing sentences are derived from cleft constructions.

Sachant que l'idee de se servir d'un verificateur de grammaire comme outil pedagogique dans le domaine de renseignement des langues secondes constitue un sujet de debat continu, nous abordons une etude d'un verificateur de grarnmaire, LINGER (Language Independent Grammatical Error Reporter), concu pour les apprenants de francais, d'espagnol et d'allemand et realise a l'Universite d'Exeter, les resultats de laquelle nous permettent de formuler des generalisations qui aident a comprendre si en theorie un verificateur de grammaire est un outil pedagogique utile ou pas. En plus, nous recommandons ce qu'il faudrait inclure dans le verificateur de grammaire ideal pour les apprenants. Afin de fournir une idee de ce dont la technologie actuelle est capable dans le domaine de la correction automatique/assistee des langues et d'etablir les merites et les problemes de l'usage d'un verificateur de grammaire comme outil pedagogique, nous presentons une etude faite sur les verificateurs de grarnmaire commerciaux, Antidote '98 et Le Correcteur 101 - Version 3.5 Professionnel. Ensuite, pour pouvoir evaluer LINGER et faire des suggestions au sujet de 1' architecture d'un systeme ideal, nous expliquons en detail les trois parties de LINGER (la grammaire, le dictionnaire et la coquille) et comment se fait l'analyse syntaxique. Avant de regarder de pres les corrections de LINGER, nous explorons une definition theorique de la correction ainsi que la facon dont l’on apprend des langues secondes. La production ecrite d'un groupe d'etudiants qui sont au niveau de la premiere annee en francais fournit les donnees avec lesquelles nous evaluons LINGER et nous nous servons d'un des verificateurs de grammaire commerciaux avec ces memes donnees afin d'evaluer les criteres de performance technique requis pour l'usage d'un tel logiciel par des apprenants. A la lumiere de tout ce que nous soulevons, nous concluons que LINGER n'est pas ni serait jamais un outil pedagogique dont des apprenants puissent se servir mais qu'en general un verificateur de grarnmaire seraient un outil pedagogique utile pourvu qu'il soit concu pour des apprenants de la meme langue maternelle; qu'il incluse des explications grarnmaticales pour chaque correction qui aident l'apprenant a mieux comprendre la regie de grammaire en question; que son analyse soit juste avec un niinimum de correction fautive, de manques d’identification de fautes et d'identification fautive de la source des fautes et qu'il fournisse une correction assistee, plutot qu'une correction automatique.

This thesis advances a specific model of 1-syntax, based on Hale and Keyser (1993, 1994) and Dechaine (1996) as a point of departure, and also proposes a general theory of the relation between the lexicon and the syntax. One of the essential proposals that I make is the F\mctionalization Principle, which permits a lexical head to project a functional projection if and only if the meaning of the head is represented by 1-syntactic structure without any extra semantic features. I refer to this type of head as a light head. The Functionalization Principle leads us to a principled account of various lexical and functional uses of lexical items such as a passive morpheme -en and have. Examples that support my analysis range from adjectival and verbal passives (e.g. Mary is very pleased and The glass was broken by BUI), to constructions of alienable and inalienable possession (e.g. John has Jive bucks and John has blue eyes), to causative/experiential constructions (e.g. John had his students walk out of class), and to perfect constructions (e.g. Lucie has advised the prime minister). Furthermore, the analysis of possessive have is extended to possessive nominals (e.g. John's cat and John's eyes). I also examine the implications of the theories of 1-syntax and 1- functors for Case. I propose that 1-syntactic structure partly determines inherent Case whereas the 1-functor checks what I call l-Junctor Case through the Spec-head relation. Furthermore, I show that these analyses of inherent Case and 1-functors account for essential properties of possessive D (a genitive marker -*s), some Hindi marked subject constructions and Japanese experiential transitive constructions.

This dissertation investigates transitivity alternations, with particular reference to Amharic. The lexical-semantic and morphosyntactic properties of morphological causatives, experiencer predicates, applicative constructions and complex predicates formed by light verbs are examined in detail. It is claimed that transitivity alternations are an artefact of Event-type alternations and follow from universal principles such as Event Headedness. It is argued that the valency difference between various verb classes reduces to whether the Root of the verb is specified or underspecified for Event Headedness. / Two levels of phrase structure, l-syntax and s-syntax, are recognized in the study. It is argued that productive causatives are generated in s-syntax, whereas morphological causatives which are sensitive to the Event-type of the Root are generated in l-syntax. A unified structural analysis is given for a number of superficially unrelated constructions including Subject Experiencer predicates, perception verbs and possessive predicates. It is argued that the quirky Case and agreement properties of such predicates can be handled by motivating inherent Case assignment. This analysis is further extended to account for the benefactive applicative of unaccusatives. / The role of light verbs in transitivity alternation is explored in detail. It is shown that light verbs are independent verbs that spell-out Event-types. The study argues that the polysemous relationship between predicates is best accounted for by a single argument structure rather than by positing multiple lexical entries.

No description available.

Code-mixing, defined as intra-sentential language alternation, is known to demonstrate structurally determined patterns of restriction. Universal constraints have been proposed to account for these structural restrictions (Poplack (1980), Woolford (1983), Di Sciullo, Muysken and Singh (1986)) but have had limited success in accounting for code-mixing between typologically diverse languages. This thesis examines the structural principles that apply universally to the interaction of languages in code-mixed sentences. We argue that systematic cross-linguistic restrictions on code-mixing can be accounted for by a syntactic constraint that is sensitive to the distinction between functional and lexical categories. / We propose the constraint of Language Indexation, according to which (structurally) adjacent categories of like functional/lexical category status must be realised in the same language. We show how this proposal accounts for code-mixed data from a range of language pairs, including Tagalog/English, Moroccan Arabic/French, Swahili/English, Irish/English, Hindi/English, Spanish/English and French/English. A difference in the application of Language Indexation in nominal versus verbal projections is discussed with reference to Tagalog/English and Moroccan Arabic/French code-mixing. Finally, we briefly examine the implications of Language Indexation with respect to the code-mixing of aphasic bilinguals.

This thesis studies the L2 acquisition of transitivity alternations in English, Spanish and Turkish within the Generative framework. in particular, it sets out to investigate the interaction of universal principles and L1 knowledge in interlanguage grammars, as well as whether similar patterns of development are observed across typologically different languages. The focus is on verbs that participate in the causative/inchoative alternation, verbs whose lexico-semantic composition---[x CAUSE [ y BECOME predicate]]---is not matched by morphological derivation uniformly across, and even within, languages. / The thesis adopts a templatic account to verb classes. Verbs are decomposed into primitive semantic predicates (CAUSE, BECOME, BE) which map onto an X-bar configuration as the heads of light verbs. Thematic roles like Agent and Theme occupy the specifiers. Different verb classes---alternating, transitive non-alternating, unaccusative and unergative verbs---derive from the interaction of semantic subpredicates and thematic roles. Transitivity alternations are determined by other narrower aspects of meaning: namely, the nature of the Agent role and the CAUSE subevent. Causative and anticausative morphology is the overt manifestation of the CAUSE and BECOME subpredicates. / Causative errors in L1 acquisition have been attributed to the overgeneralization of the causative/inchoative alternation to unergative and unaccusative verbs. This study proposes that these errors result when children incorrectly map non-alternating transitive and intransitive verbs onto a default transitive template. By extending this proposal to the L2 acquisition situation, it is hypothesized that L2 learners of Turkish, Spanish and English of different language backgrounds and at lower proficiency levels also rely on this universal mechanism when learning transitivity alternations; L1 influence only plays a role with the overt/non-overt morphology of alternating verbs in these languages. / Three independent but methodologically identical experimental studies on English, Spanish and Turkish as second languages are presented. Overall, results of a Picture Judgment Task confirm the hypotheses in the three studies. This work argues against the Full Transfer/Full Access Hypothesis (Schwartz & Sprouse 1996) as a theory of L2 competence and advances a modular view of Transfer which proposes that L1 influence does not affect all linguistic domains in the same way.

The word pun is one of the figures of speech that people employ in everyday communication and especially in literary works in order to advance intricate aspects of meaning that may not be easy to express using 'plain' language. The word pun generally provides a speaker with an opportunity to mean more while saying less instead of saying more while meaning less. Considering the facts that people primarily communicate in order to exchange meaning, and that meaning can be very elusive and controversial, there are two questions about the use of the word pun. Can people understand the meaning in word puns? And, if they can, how do they do so? These questions are especially relevant in the case of literary works because the author is far removed from the audience. As such, there is no room for the negotiation of meaning. The current study probes these two questions by considering two types of word pun, Chiasmus and Metathesis, which are composed through the transposition of the morphological and or syntactic order of expressions. At a theoretical level, the study explores and explains the common underlying processes that guide the comprehension of the word puns. Further than that, it demonstrates through a case study that people are able to understand the word puns by using what the study proposes to be the reverse principle. Ultimately, the study illustrates how people derive meaning of utterances through interplay of the different subsystems of the language system. The word puns in this study present a good context in which to explore interfaces between semantics and other language fields by linking insights drawn from different fields of linguistic enquiry to the concept of meaning and demonstrating how aspects of all these come together in explaining the processing of word puns that involve morphological and syntactic transposition. Additionally, the study demonstrates that people understand the world by relating concepts to one another because of the underlying relationships existing between concepts and by virtue of the relationships that hold between and among words or word parts. / Thesis (M.A.)-University of KwaZulu-Natal, 2007.

AIM The aim of this study is to investigate the opacity of Chinese compounds in search of conceptual motivations for traditional exocentric compounds and contemporary neologisms in Chinese. METHODOLOGY This research may be characterised as an empirical investigation within the quantitative paradigm. The study contains three tasks. The design of Task 1 and Task 2 replicates the experiment concerning the classification of compound transparency which Libben, Gibson, Yoon and Sandra (2003) used to test English compounds. Task 3 is a kind of word association task that is designed following a suggestion by Gleason and Ratner (1998: 215). A sample of 95 Chinese native speakers for Task 1 & Task 2 is used. A sample of 50 Chinese native speakers for Task 3 is used. None of them has participated in either Task 1 or Task 2. FINDINGS The findings are presented with regard to the two types of compounds investigated in the study: ‘semantically free’ compounds and neologisms. In summary, ‘semantically free’ compounds may process through their constituents in the mental lexicon. Meanwhile, for some certain reasons ‘semantically free’ compounds may be recognized from the mental lexicon as whole. In the research, it found that the frequency effect is stronger than the effect of ‘semantic transparency’ in ‘semantically free’ compounds, it could mean that lexico-semantic distance (semantic freedom) is much smaller in Chinese exocentric compounds than anticipated by Scalise and Guevara (2006). Neologisms may process through their constituents in the mental lexicon. The effect of semantic transparency may be stronger than the frequency effect in neologisms when compounds are semantically transparent and their constituents’ meanings are similarity. KEY CONCEPTS Exocentric compounds, endocentric compounds, ‘semantically free’ compounds, neologisms, opaqueness, semantic transparency, frequency effect, word-superiority effect. / Thesis (M.A.)-University of KwaZulu-Natal, Durban, 2011.