Inaug.-Diss.--Breslau. / Lebenslauf. "Literatur": p. 87-93.

Thesis (Doctoral)--Königliche Universität Greifswald, 1916.

Thesis (MA)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: This study deals with the phenomenon of obligatory reflexivity in Chichewa, a language belonging to the Bantu family. Obligatory reflexivity occurs in constructions where a pronominal element – e.g. the reflexive marker (RFM) -dzi- in the verbal complex in Chichewa – is referentially dependent on some other expression in the sentence, its antecedent. Constructions of this type have not received systematic attention in the literature on Chichewa syntax, except in the works of Mchombo (1993, 2004, 2007). The first objective of the study is to fill this empirical gap by providing a detailed description of the different types of construction in which obligatory reflexivity is found in Chichewa. The second objective is to provide an analysis of the relevant facts within the broad framework of Minimalist Syntax (e.g. Chomsky 1995, 2000; Hornstein, Nunes & Grohmann 2005; Radford 2009). More specifically, the study seeks to determine whether the recent minimalist approach to the analysis of obligatory reflexive constructions put forward by Oosthuizen (2013), the so-called Nominal Shell Analysis (NSA), provides an adequate framework for analysing reflexive constructions in Chichewa. The analysis that is set out in this study focuses on three types of reflexive construction, namely verbal object reflexives, infinitival verbal reflexives, and infinitival nominal reflexives. It is argued that an analysis that incorporates the core hypotheses and devices of the NSA can provide a proper description and explanation of the facts of obligatory reflexivity as reflected in these three types of construction. In particular, it is claimed that such an analysis can account for the establishment of a coreferential relationship between the RFM -dzi- and an antecedent, without requiring any special devices or devices that are incompatible with the basic assumptions of the minimalist approach to linguistic inquiry. In brief, according to the analysis, the RFM -dzi- and its antecedent are initially merged into a light noun phrase, nP, with the RFM representing the functional n-head of this phrase. In this configuration, the coreferential relationship between the antecedent and -dzi- is established when the antecedent provides the RFM with φ-values (i.e. values for the grammatical features person, number and noun class). In the course of the discussion, several proposals are also put forward in connection with other, related aspects of Chichewa syntax, including the agreement relationship between the subject/object and their respective markers in the verbal complex. / AFRIKAANSE OPSOMMING: Hierdie studie handel oor die verskynsel van verpligte refleksiwiteit in Chichewa, ’n lid van die Bantoe-taalfamilie. Verpligte refleksiwiteit kom voor in konstruksies waar ’n pronominale element – bv. die refleksiefmerker (RFM) -dzi- in die verbale kompleks in Chichewa – referensieel afhanklik is van ’n ander uitdrukking in die sin, die antesedent. Konstruksies van hierdie tipe het nog nie sistematies aandag gekry in die literatuur oor Chichewa sintaksis nie, behalwe in die werke van Mchombo (1993, 2004, 2007). Die eerste hoofoogmerk van die studie is om hierdie empiriese gaping te vul deur ’n gedetailleerde beskrywing te gee van die verskillende tipes konstruksie waarin verpligte refleksiwiteit in Chichewa aangetref word. Die tweede hoofoogmerk is om ’n analise te gee van die tersaaklike feite binne die breë raamwerk van Minimalistiese Sintaksis (bv. Chomsky 1995, 2000; Hornstein, Nunes & Grohmann 2005; Radford 2009). In meer spesifieke terme word daar nagegaan of die minimalistiese benadering tot die analise van verpligte refleksiwiteit wat onlangs voorgestel is deur Oosthuizen (2013), die sogenaamde Nominale Skulp-analise (NSA), ’n toereikende raamwerk bied vir die analise van refleksiefkonstruksies in Chichewa. Die analise wat uiteengesit word in hierdie studie fokus op drie tipes refleksiefkonstruksie, naamlik verbale objek-refleksiewe, infinitiewe verbale refleksiewe, en infinitiewe nominale refleksiewe. Daar word geargumenteer dat ’n analise wat gebruik maak van die kernhipoteses en meganismes van die NSA ’n behoorlike beskrywing en verklaring kan bied van die feite van verpligte refleksiwiteit soos dit voorkom in hierdie drie tipes konstruksie. In besonder kan so ’n analise ’n beskrywing en verklaring gee van die manier waarop ’n koreferensiële verhouding tussen die RFM -dzi- en ’n antesedent bewerkstellig word, sonder die nodigheid van spesiale nuwe meganismes of meganismes wat onversoenbaar is met die basiese aannames van die minimalistiese benadering tot taalondersoek. Die analise hou kortliks in dat die RFM -dzi- en sy antesedent aanvanklik saamgevoeg word in ’n ligte naamwoordfrase, nP, met die RFM wat optree as die funksionele n-hoof van hierdie frase. In dié konfigurasie word die koreferensiële verhouding tussen die antesedent en -dzi- bewerkstellig wanneer die antesedent die RFM van φ-waardes voorsien (d.i. waardes vir die grammatikale kenmerke persoon, getal en naamwoordklas). In die loop van die bespreking word daar ook verskeie voorstelle gemaak oor ander, verwante aspekte van Chichewa sintaksis, onder meer oor die kongruensie-verhouding tussen die subjek/objek en hulle onderskeie merkers in die verbale kompleks.

Querying a database to retrieve an answer, telling a robot to perform an action, or teaching a computer to play a game are tasks requiring communication with machines in a language interpretable by them. Semantic parsing is the task of converting human language to a machine interpretable language. While human languages are sequential in nature with latent structures, machine interpretable languages are formal with explicit structures. The computational linguistics community have created several treebanks to understand the formal syntactic structures of human languages. In this thesis, we use these to obtain formal meaning representations of languages, and learn computational models to convert these meaning representations to the target machine representation. Our goal is to evaluate if existing treebank syntactic representations are useful for semantic parsing. Existing semantic parsing methods mainly learn domain-specific grammars which can parse human languages to machine representation directly. We deviate from this trend and make use of general-purpose syntactic grammar to help in semantic parsing. We use two syntactic representations: Combinatory Categorial Grammar (CCG) and dependency syntax. CCG has a well established theory on deriving meaning representations from its syntactic derivations. But there are no CCG treebanks for many languages since these are difficult to annotate. In contrast, dependencies are easy to annotate and have many treebanks. However, dependencies do not have a well established theory for deriving meaning representations. In this thesis, we propose novel theories for deriving meaning representations from dependencies. Our evaluation task is question answering on a knowledge base. Given a question, our goal is to answer it on the knowledge base by converting the question to an executable query. We use Freebase, the knowledge source behind Google’s search engine, as our knowledge base. Freebase contains millions of real world facts represented in a graphical format. Inspired from the Freebase structure, we formulate semantic parsing as a graph matching problem, i.e., given a natural language sentence, we convert it into a graph structure from the meaning representation obtained from syntax, and find the subgraph of Freebase that best matches the natural language graph. Our experiments on Free917, WebQuestions and GraphQuestions semantic parsing datasets conclude that general-purpose syntax is more useful for semantic parsing than induced task-specific syntax and syntax-agnostic representations.

abstract: The primary topic of this dissertation is the grammaticalization of negation in three Sinitic language varieties: Hakka, Mandarin, and Southern Min. I discuss negative morphemes that are used under different modality or aspect contexts, including ability, volition, necessity, and perfectivity. Not only does this study examine Southern Min affirmative and negative pairs, but it also highlights the grammaticalization of negation and parametric differences in negation among the languages under investigation. This dissertation also covers the reanalysis of negatives into interrogatives. I approach the investigation of Southern Min negation from both synchronic and diachronic perspectives. I analyze corpus data in addition to data collected from fieldwork for the contemporary linguistic data. For my diachronic research of Chinese negation, I use historical texts and etymological dictionaries. Diachronically, many of the negative morphemes originate from full-fledged verbs and undergo an analogous grammaticalization process that consists of multiple stages of reanalysis from V to T (aspect; modality), and then T to C (interrogative; discourse). I explain this reanalysis, which involves head-to-head movement, using generative frameworks that combine a modified cartographic approach and the Minimalist Economy Principles. Synchronic data show that Southern Min affirmative modals are characterized by a certain morphological doubling. These doublings consist of two near synonyms used in sequence, resulting from the loss of features in a verb and a second verb added as a renewal. In the negation paradigm, some negatives project a negative phrase, while the others serve a dual function, occupying a modal/aspect head as well as a negative head. The latter system is gradually shifting to the former. This study uncovers evidence to counter the long-established paradigm, where negation is tied to its independent modality (abilitive, volitional and necessitive) or aspect (perfective and perfect). I observe a mismatch between the use of interrogatives and their modality/aspect and attribute this phenomenon to feature loss during their reanalysis from negatives to interrogatives. Results however show that consistency occurs in the grammaticalization of negation within Southern Min and intra-linguistically among the three Sinitic languages, and that parametric differences are found at the morphological level. / Dissertation/Thesis / Ph.D. English 2012

Recent research in both native (L1) and non-native (L2) knowledge of quantifier scope has led to a number of competing beliefs about the nature of learner knowledge. With regard to native knowledge, it has been noted in the literature that there is a discrepancy between L1 child and adult performance in quantifier interpretation. This observed mismatch has led to the formulation of two conflicting analyses of the L1 data. Philip (1991, 1992, and 1995) and others (Philip and Takahashi, 1991; Roeper, Strauss and Pearson, 2004, 2005; DelliCarpini, 2003) propose that quantification is a natural acquisition process constrained by Universal Grammar (UG) in which children progressively mature in their competence until they converge upon an adult grammar. Conversely, Crain (1995, 1996, and 1998) and others (Musolino, Crain and Thornton, 2000; Musolino and Lidz, 2006) maintain that children as young as five years old have a mature competence and that the failure to apply semantic principles is the result of the infelicitous nature of experimental task items. Essentially, the former account posits imperfect child L1 competence, while the latter asserts perfect competence. Similar research in the non-native (L2) knowledge of quantifier scope has been motivated by two essential questions: 1) Is adult L2 acquisition constrained by the same innate linguistic mechanisms as L1 acquisition; and 2) what is the role of L1 knowledge in adult L2 acquisition? (Marsden, 2004b: 9). In consideration of these questions, two main approaches have been devised as predictive models (following Epstein, Flynn and Martohardjono, 1996; Grüter, Lieberman and Gualmini, 2008, 2010). Under the Full Transfer hypothesis, the learner is predicted to approach an L2 with the same values, settings and preferences of the L1, whereas under the Full Access account, the L2 learner is informed by the Language Faculty directly without the intervening effects of the L1 (Grüter et al., 2008: 47). A third approach, Schwartz and Sprouse's (1996) model, unites both Full Access and Full Transfer to explain L2 acquisition. The present study explores L1 Mandarin knowledge of L2 English quantifier scope in order to address the issue of perfect vs. imperfect competence (as applied to SLA), as well as the matter of access to UG vs. L1 transfer. Incorporating insights from DelliCarpini (2003), I first assess the presence of symmetric and exhaustive responses, which are indicative of an immature grammar. Crucially, I use one group (n=11) of L2 English speakers (low intermediate and advanced) in order to test for a maturational discrepancy that would putatively differ according to proficiency. Secondly, I identify a potential poverty of the stimulus situation: L1 transfer cannot account for the L2 acquisition of English non-isomorphic scope licensing by native speakers of Mandarin Chinese due to the Isomorphic Principle (Aoun and Li, 1993). If demonstrated, the ability of L2 learners to converge upon this semantic principle (non-isomorphism) will serve as potential evidence for access to Universal Grammar in adult non-native learners.

No description available.

This thesis is primarily concerned with the developments that have taken place in infinitival constructions in the late Old English, Middle English, and Modern English periods. It is an investigation into the status of Old English to-infinitive, the origin, nature, and distribution offor in Middle English (for)to-infinitival constructions, and the origin and reanalysis offor as a complementiser in the Modern English [for DP to VP] construction. In chapter one, we introduce some of the basic notions of the Minimalist Program outlined in Chornsky (1993,1995)). The airn of chapter two is to account for the structural status of to in the Old English toinfmitive. It is argued in this chapter that firstly (functional) C, Agr, and T are not eligible positions for to, and secondly that to occupies the lexical category P(reposition). The prepositional status of the Old English to-infinitive is supported by the fact that it occurs in coordination with ordinary PPs. Chapter three argues that the Old English to-infinitive should be treated as a single (morphological and) syntactic unit which cannot be broken up by intervening elements. We propose that to is generated with a D-feature and that the infinitival verb is a combination of two features: an Inf-feature and a D-feature. We argue that as long as V+W-to-D movement is attested, the syntactic unity cannot be broken up by elements like objects, adverbs, etc. Once the Old English case system disintegrated, the internal structure of the to-infinitive underwent a radical change such that the demise of -ne (which resulted from the weakening of to as a dative case-assigneri resulted in the demise of D, and this led to the disintegration of the syntactic unity of the to-infinitive, and the consequent appearance of for before to. In other words, when to ceased to be a preposition, for moved in and 'took over' (and perhaps became an infinitival marker as well, givingforto). In chapter four, we proceed to account for the structural status offor in Middle English to-infinitive. Three analyses that attempt to account for the status of for are examined and rebutted in favour of our analysis of for as part of the infinitival morphology. Chapter five provides morphological and syntactic evidence in favour of analysingfor and to as a compound infinitival marker. It is argued that the position of the compound infinitival marker (for)to is T(ense). This analysis correctly predicts (for)to to be present in raising and control infinitives. A number of factors which show that (for)to occupies T will be noted and discussed. The purpose of chapter six is to provide evidence for the correlation between verb movement and object shift in Middle English (for)to-infinitives. It will be argued that the infinitival verb moves overtly from VP to Inf, the functional head which hosts the infinitival feature. Some empirical evidence relating to conjoined structures and VPadverbs is discussed. The attestation of V-to-Inf movement in Middle English (for)toinfinitives is strongly supported by the presence of object shift. Our conclusion is that the non-attestation of object shift in Modern English to-infinitives can be attributed to the absence of overt V-to-Inf movement. Having established the morphological and syntactic status of the infinitival marker (forfto (chapter five) and the infinitival verb (chapter six), we proceed to investigate the origin offor in the Modem English [for DP to VPJ construction. On the basis of morphological and structural evidence, we propose that the [for DP to VP] construction is the outcome of two diachronic reanalyses (DRs), which took place at two different stages in the history of English. The first DR, which took place in the 12th century, was triggered by the loss of dative case which paved the way for the introduction of prepositions like for to realise the benefactive function. In Old English the benefactive function was typically associated with morphological dative case. Once dative case had been lost, the beriefactive. function had to be realised by prepositions likefor. Throughout the Middle English periodfor was a case-realiser and not a lexical preposition. Its main function was to realise an inherent case feature which belonged to the matrix lexical head. The second DR, which occurred in the 16th century, was triggered by the fact that the string [for DP to VP] had become structurally ambiguous for acquirers, allowing an interpretation where [for DPI is part of the matrix predicate, or alternatively an interpretation where [for DPI is the subject of the infinitival clause. In the latter interpretation for's function is to realise a Case which does not belong to any lexical head. It realises the Case property of the C-position. It will be argued that the prepositionfor was reanalysed as a complementiser as a result of the loss of infinitival clauses as complements of prepositions, and the consequent development of the C-position as a potential accusative Case licenser. The change can be regarded as a change in the status offor from a lexical case-realiser to a functional Case-realiser.

abstract: This study explores the acquisition of the determiner phrase (DP) in monolingual (L1) and bilingual (2L1) French. I investigate the acquisition of DP structures and features in the speech of two monolingual French and two bilingual French-English subjects from the CHILDES (Child Language Data Exchange System) corpus. I perform a thorough, longitudinal examination of the children's data, from the ages of 1;10 to 4;00, focusing on the description and analysis of their development of DP elements, words, and structures such as the definite and indefinite articles, demonstratives, and numerals, as well as the DP features of gender, number, and definiteness. I also consider the Adjective Phrase (AP) and its interaction with the DP. This study complicates the traditional view of discrete, simplified stages of DP acquisition, arguing instead for an ongoing and complex process. Application of the Minimalist model of syntactic analysis provides essential insights into the underlying processes of child grammar, and suggests a number of previously unaddressed characteristics and patterns in French DP development. / Dissertation/Thesis / Doctoral Dissertation English 2015

abstract: This study explores the development of negation and the Negation Phrase (NegP) in bilingual children learning both English and Spanish. I analyze the speech of four children growing up in the United States who are learning English and Spanish simultaneously in order to establish steps of parameter setting for negation. The transcripts have been taken from Pérez-Bazán’s bilingual corpus from CHILDES (Child Language Data Exchange System). The thorough analysis of the selected corpus data gathered from children ages 2;0 and 3;3 determines the steps children follow in order to develop mastery of negation and the NegP. This study is an addition to the body of research surrounding language acquisition and the concept of Universal Grammar’s Principles and Parameters framework. The bases for this study is Klima & Bellugi’s (1968) established three steps for acquisition of negation by children in English, as well as Zeijlstra’s (2004) analysis of languages in regards to phases of the Jespersen cycle. The data of this study suggest that there are five basic steps to parameter setting, and that as utterances become syntactically more complex, children value uninterpretable features with interpretable ones. This is seen in both languages studied. The parameters categorized based on the data available for this study are the following: 1) negative particle outside of the VP, 2) NegP creation and development with preverbal negative marker, 3) Negative Concord (NC), 4) True Imperatives (banned or not), and 5) Negative Polarity Items (NPI). Also important is the placement of the NegP, as it is above the TP in Spanish and c-commanded by the TP in English. The development of the NegP and uninterpretable negation [uNeg] valuation by interpretable negation [iNeg] is also explored in the utterances of the four children studied. This study confirms Klima & Bellugi’s account of steps and further defines child negation in English as well as in Spanish. The focus on [iNeg] and [uNeg] features is further explained using Zeijlstra’s Phases of the Jespersen cycle as a springboard. I add salient information regarding parameter setting and how negation and the NegP are developed across two languages. / Dissertation/Thesis / Doctoral Dissertation English 2015