Doctoral Degrees (General Linguistics)

Permanent URI for this collection

https://scholar.sun.ac.za/handle/10019.1/611

Browse

Browsing Doctoral Degrees (General Linguistics) by browse.metadata.advisor "Biberauer, Theresa"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Nominal classification in Bantu revisited : the perspective from Chichewa
    (Stellenbosch : Stellenbosch University, 2019-12) Msaka, Peter Kondwani; Biberauer, Theresa; Stellenbosch University. Faculty of Arts and Social Sciences. Dept. of General Linguistics.
    ENGLISH ABSTRACT: The present study focuses on the phenomenon of noun class and agreement systems in Chichewa, a Bantu language spoken in some parts of Eastern and Southern Africa. Noun classification is a strong areal feature of Bantu languages, however it remains one of the less well understood phenomena in this family (see i.a. Maho, 1999, chap. 6; Mchombo, 2004: 3; Katamba, 2006: 120). The generally accepted views about this phenomenon centre on the noun-class framework associated with Bleek (1862, 1869) and Meinhof (1899, 1906, 1932) – henceforth referred to as the Bleek-Meinhof system. The major problem with this system lies in the nature of the assumptions that are made regarding the principles underlying the classification system. In terms of the Bleek-Meinhof system, it is assumed that modern-day Bantu noun-class systems can all be understood as being based on a set of reconstructed noun prefixes. If we consider these modern-day systems, however, it emerges that these noun prefixes are not found on every noun and also not in every noun class. This is a distribution pattern that has been argued to have arisen due to language change, which has distorted the original regular morphological and semantic basis of the classes. The present study is centrally concerned with Chichewa, a language whose present-day noun class system poses numerous challenges to the traditional Bleek-Meinhof system. I argue that this latter system can in fact be shown to have been constructed on the basis of a partial dataset. Drawing on a carefully sampled dataset from the Chichewa monolingual dictionary, I outline how many nominal classifications can be shown to be problematic if we adopt the Bleek-Meinhof system. Clearly setting out these facts, constitutes the thesis’ descriptive objective. With the description in place, the second goal is to propose a novel noun-class system that is plausible from a child language acquisition perspective. I propose an agreement-based approach which classifies all expressions in the dataset according to the agreement markers that they control on adjectives, numerals, verbs, etc. Having taken this approach, I demonstrate that we can identify twelve agreement classes (ACs) in Chichewa, with each AC including a wide range of expressions, some of which are non-nominal in nature. In this regard, I propose two subsystems within the agreement classification system, one an agreement-based noun-class (ANC) subsystem and the other a general agreement class (GAC) subsystem. On the one hand, I will argue that in Chichewa, the ANC system has a primarily semantic basis, which gives rise to two super-noun classes, one animate/agentive and the other non-agentive/inanimate. The inanimate/non-agentive is further subcategorised on a phonological basis, taking into account the word-initial element, into six other classes. The GAC, on the other hand, is triggered by on at first sight rather miscellaneous group of expressions, some of which have not traditionally been associated with nominal expressions such as complementiser phrases (CPs), prepositional phrases (PPs), and also underspecified agreement triggers. I also argue that the proposed underlying principles of the Chichewa ANC and GAC systems are at the core of various grammatical structures and language acquisition processes observed crosslinguistically.
  • Thumbnail Image
    Item
    Obligatory reflexivity in Afrikaans : a minimalist approach
    (Stellenbosch : Stellenbosch University, 2013-03) Oosthuizen, Johan; Biberauer, Theresa; Stellenbosch University. Faculty of Arts and Social Sciences. Dept. of Philosophy.
    ENGLISH ABSTRACT: This study focuses on the phenomenon of obligatory reflexivity in Afrikaans. Despite a considerable literature on this phenomenon as it is reflected in other languages, the Afrikaans data have not received any systematic attention. Hence, a first major aim is to address this empirical gap. Secondly, informed by the Afrikaans data, the study aims to develop an analysis that can provide a conceptually adequate account for the facts, and that is amenable to extension beyond Afrikaans. The proposed nominal shell analysis (of obligatory reflexivity) (NSA) is developed within, on the one hand, the general framework of Minimalist Syntax and, on the other hand, the specific framework of proposals about word order and linearisation phenomena in Germanic languages worked out in, amongst others, Holmberg (2000), Biberauer (2003), Biberauer & Richards (2006), Biberauer & Roberts (2006), and Biberauer et al. (2009, 2011). The basic idea underlying the NSA is that two expressions which enter into an obligatory coreferential relationship are initially merged into a nominal shell structure headed by an identity focus light noun n. It is argued that the identity focus n belongs to a natural class of identificational elements which also includes a contrastive focus n, a presentational focus n, a possessor focus n, and a quantity focus n. In terms of the NSA, the identity focus n takes a reflexive pronoun as its complement, with such a pronoun being analysed as a syntactic compound that is derived by merging a category-neutral lexical root √PRON with a D constituent containing unvalued φ-features. This means, then, that a reflexive pronoun is defined in syntactic terms and not in terms of special lexical features. The reflexive is subsequently raised to the identity focus n – which forms the locus of the suffix -self associated with morphologically complex reflexive pronouns – where it is spelled out as part of the compound n that is derived in this manner. The antecedent expression is next merged as the specifier of the compound light noun, resulting in a configuration where the antecedent can value the φ-features of the reflexive, with the n serving as mediator. In this configuration, the φ-valued pronoun is semantically interpreted as an anaphor and the nominal expression in the specifier position of the nP as its antecedent; that is, the pronoun is interpreted as obligatorily coreferential with this nominal expression. The details of the NSA and its empirical and conceptual consequences are worked out with reference to six constructions in which reflexive pronouns can occur: verbal object constructions, prepositional object constructions, double object constructions, infinitival constructions, small clause constructions, and possessive constructions. Brief attention is also given to the possibility of extending the ideas underlying the NSA to (i) languages of the Southern Bantu family, where the reflexive element surfaces as a verbal affix, and (ii) two further types of construction in Afrikaans which seem amenable to such a nominal shell approach, namely floating quantifier constructions and expletive daar (“there”) constructions.
  • Thumbnail Image
    Item
    Predictive variables in lifelong bilingualism : an exploratory study probing the effects of L2 English on L1 Afrikaans syntax
    (Stellenbosch : Stellenbosch University, 2022-12) Van Heukelum, Marie-Louise; Biberauer, Theresa; Bylund, Emanuel; Stellenbosch University. Faculty of Arts and Social Sciences. Dept. of General Linguistics.
    ENGLISH ABSTRACT: This exploratory study is concerned with effects of the second language on the first (EotSLotF). It specifically aims to probe the extent to which it is possible to establish whether and, if so, how the morphosyntactic system of Afrikaans as a first language (L1) changes under the influence of English as a second language (L2) as a result of language exposure and use across the lifespan. Research focussing on EotSLotF is typically concerned with either heritage language (HL) development or L1 attrition. The present study investigates variables typical of both HL development (i.e. exposure/use in childhood/adolescence) and L1 attrition (i.e. exposure/use in adulthood), thus bringing these two sub-fields together in an attempt to deepen our understanding of how L1 Afrikaans develops under the influence of L2 English across the lifespan. Tsimpli (2014) argues for a three-way distinction in L1 grammatical development between early, late and very late properties: early properties turn on narrow syntax, while late and very late properties require syntax- and potentially language-external mapping. Similarly, studies of L1 attrition show crucial differences between early/narrow-syntax and late/interface- driven properties: the former are suggested to be less vulnerable to attrition as they incur fewer processing-related challenges. This observation is central to the so-called Interface Hypothesis (IH; Sorace & Filiaci, 2006 et seq.). This study tests the early-acquired/interface-internal versus late-acquired/interface-external asymmetry in Afrikaans-English bilinguals, a language combination and population not previously investigated within the IH framework. 166 participants in South Africa and the diaspora completed (i) a language background questionnaire, (ii) an acceptability judgement task (AJT), and (iii) a contextualised acceptability judgement task (CAJT). Five syntactic properties of Afrikaans, which differ with respect to their sensitive periods and their relationship to the interfaces, were investigated: (i) Verb Second (V2), (ii) basic sentential negation, (iii) double negation (DN), (iv) pronominal scrambling, and (v) discourse-driven scrambling. The study also takes into account sociolinguistic variation: both what is prescriptively sanctioned in Standard Afrikaans (StdA) and what is permissible in Modern Spoken Afrikaans (MsA) is investigated. The results reveal that, overall, variation indicative of EotSLotF appears to be minimal in the population under investigation. The earlier-acquired properties of the narrow syntax/internal interfaces, while not impervious to the effects of differing amounts of L1/L2 exposure and use, in particular exhibit remarkable stability. The MsA judgement patterns suggest that more frequent exposure to and use of Afrikaans is facilitative in predicting “target- like” behaviour. Where variation indicative of EotSLotF is evidenced, the patterns are, in some cases, similar to those seen in traditional HS populations. The results also suggest, however, that language-specific sociolinguistic and language-internal factors may be centrally relevant. The picture that emerges is that the IH is a useful framework for probing L1 (in)stability under the influence of an L2 in populations resembling the Afrikaans-English type. However, beyond sensitive-period and interface considerations, language-specific factors appear to play a non-trivial role in the more fine-grained shaping of the L1 grammar. These factors therefore appear to warrant special attention in research concerned with EotSLotF.
  • Thumbnail Image
    Item
    Spelling out P: a unified syntax of Afrikaans adpositions and V-particles
    (Stellenbosch : Stellenbosch University, 2017-12) Pretorius, Erin; Biberauer, Theresa; Corver, Norbert; Oosthuizen, Johan; Van Koppen, Marjo; Stellenbosch University. Faculty of Arts and Social Sciences. Dept. of General Linguistics.
    ENGLISH ABSTRACT: Elements of language that are typically considered to have P (i.e. adpositional) category status frequently exhibit divergent morphosyntactic properties, and it is often the case that one and the same element exhibits divergent morphosyntactic properties. Such elements are syncretic. An important fact about syncretism is that it poses a challenge to the ontologically primitive syntactic category. With a concentrated focus on the Afrikaans spatial P domain, this dissertation develops a system in which observed patterns of syncretism fall out naturally from (i) the fine-grained cartographic structure of the non-primitive P domain, (ii) the “shape” of the formal featural specification on particular (classes of) P elements, and (iii) a theory of how lexical material is matched and inserted to express syntactic structure. In this system, syntactic categories are not ontologically primitive but are composite syntactic objects consisting of (overlapping) sets of hierarchically structured formal features. Category effects – all the morphosyntactic characteristics associated with a particular category – arise as epiphenomena of the particular set of features that an element lexicalises at a particular insertion site. As the book progresses, it is demonstrated how all the language-internal variation in expressions containing P elements – simplex and complex prepositional phrases, circumpositional phrases, doubling adpositional phrases, and particle verbs with P-based particles – can be derived from the same basic structure. On the proposed analysis, category boundaries are non-discreet and may be spanned by individual lexical items, accounting for the multiple macro-category membership of some P elements using precisely the same mechanisms that account for multiple micro-category membership.