Thesis (Ph. D.)--Hanʼguk Oegugŏ Taehakkyo, 1990. / "August 1990." "8"--Spine. Abstract in Korean. Includes bibliographical references (p. 167-173).

Thesis (Ph. D.)--Temple University, 1988. / Includes bibliographical references.

Thesis (Ph. D.)--Cornell University, 1985. / Vita. Includes bibliographical references (p. 257-261).

Magister Artium - MA / The purpose of this study is to investigate how subject PRO of the infinitive in Sesotho is assigned an antecedent The general framework which is assumed is the Government Binding theory of generative grammar, in particular one of its subtheories, namely the Control theory which is concerned with the assignment of an antecedent to the subject PRO in the infinitive. A survey of the work done by various linguists on the Infinitive in Bantu languages shows that much has been written on the properties of the infinitive. In the Nguni languages, especially in Xhosa, considerable research has been done to establish the syntactic nature of the infinitive in the various constructions that it may occur. The Infinitive in Sesotho has, however, not been studied extensively. The central issue of this study relates to establishing the antecedent of the subject PRO of the infinitive in the various constructions in which it may appear. It is demonstrated that in some instances it is possible for two different NP arguments in a sentence to serve as a possible antecedent of the empty category PRO. This examination of control of the subject PRO in the infinitive in Sesotho has revealed the following: The subject PRO in the infinitive must have antecedent. In such cases the empty category PRO which is obligatory control is considered to behave like an anaphor, because as an empty of the clausal complement it must take its referential index from either the subject or object argument of the matrix sentence. This implies that the empty category PRO has no capacity for independent reference. 2. The subject PRO in the infinitive may have an In such cases the empty category PRO which is subject to non-obligatory control is considered to behave like a pronoun, because the empty category PRO may either refer to individuals independently or co-refer to individuals already named on a given sentence. Finally the subject PRO in the infinitive can at times have no antecedent at all In such cases the empty category PRO is subject to arbitrary control, because its antecedent may be implicit This has been established in the following instances: 3.1 There are cases where the infinitive is a complement of a Verb or a Copulative with the subject position occupied by an empty existential pronominal which is associated with the existential morpheme Ho. This pronominal is an empty category just like PRO, but is in a non-argument position, for it lacks a thetha-role. This is attributed to the fact that the existential pronominal associated with Ho, which signifies "it" or "there" is a dummy element and has therefore no semantic role. Therefore the antecedent of the subject PRO in the infinitive can be "anyone in general". The interpretation thereof is known as arbitrary control. It has also been established that, the Nominal infinitive, which has in Sesotho, like all other nominals, a class prefix which is morphologically marked with a prefix Ho, is subject to arbitrary control (see 3 above). This is attributed to the fact that its subject NP-argument has a non-referential expression which is thus ungoverned, because it has no agreement (-AGR) and can therefore not be assigned Case.

No description available.

This study applies Chomsky's Barriers version of government binding theory to subject clitic pronouns and subject-verb agreement processes in Somali. It is proposed that Somali has a contrast between strong and weak subject-verb agreement, indicated as AGR$ sb{ rm S}$ and AGR$ sb{ rm W}$ respectively. SCL is possible iff AGR = AGR$ sb{ rm S}.$ SCL is obligatory if subject is pro, the null pronominal. SCL is excluded in case of short subject extraction, which is only possible from the domain of AGR$ sb{ rm W}.$ The prohibition against short subject extraction from the domain of AGR$ sb{ rm S}$ is attributed to Principle B of Aoun's generalized binding theory. Long subject extraction from the domain of AGR$ sb{ rm S}$ is possible if an intermediate trace in (NP,CP) deletes before binding theory. The analysis is extended to subject extraction in Italian, including the Trentino dialect, and Modern Irish.

"A syntactic Analysis of Kinyarwanda applicatives" is a study of the syntax of Kinyarwanda which focuses on applicatives. Applicatives are constructions in which the object of a preposition becomes the direct object of the verb through a grammatical function changing process. In such constructions, the verb bears a morpheme referred to as the applicative morpheme which turns an intransitive verb into a transitive verb and a transitive verb into a ditransitive verb. The derived object may perform various thematic functions, including those of instrument, beneficiary, goal, manner, reason, purpose and motive. The study provides a thorough description of different types of ditransitive applicatives in Kinyarwanda by examining the syntactic properties exhibited by both objects. In general Kinyarwanda may be classified as a 'symmetrical' language in which more than one object can exhibit direct object properties. This is true for instrumental, benefactive and manner applicatives. However, some applicatives in Kinyarwanda such as the locative applicatives are 'asymmetrical' in that only one object exhibits all the direct object properties. In my research I analyse Kinyarwanda applicatives within the framework of Principles-and-Parameters (Chomsky 1981, 1986a, b and subsequent work), more specifically the Government and Binding theory (Chomsky 1981). I base the discussion on three analyses that have been proposed in the literature of applicatives: Baker's (1988) preposition incorporation theory, Larson's (1988) double object construction analysis and Nakamura's (1997) account of object extraction in applicative constructions, which is based on Chomsky's (1995) Minimalist Program. The study shows that the above analyses account for some aspects of applicatives, but that there are certain facts that are not accounted for, which require a different analysis. / Thesis (M.A.)-University of KwaZulu-Natal, 2005.

This dissertation investigates the structure of Internally Headed Relative Clauses (henceforth, IHRCs) within the framework of Government and Binding Theory. I propose that the structure of IHRCs involves a modifying clause predicted of a head. We provide empirical as well as conceptual arguments showing that the head of an IHRC is a pro licensed by coindexation with the internal head. Following Safir's (1986) LCR, this licensing must be mediated by A$ sp prime$-Binding, which is effected by operator-movement to COMP within the modifying clause. Island Constraints provide evidence for operator-movement in IHRCs. IHRCs pattern with other Wh-constructions in this way. Correlatively, if there are no Island effects in IHRCs, there are no Island effects with other Wh-constructions (cf. Lakhota, Mojave). In this case an Unselective Binding relation is established in the language. Other properties such as the Definiteness Effect and the occurence of Switch-reference markers in IHRCs in Choctaw are also shown to follow from independent parameters. / The consequences of our analysis for Baker's (to appear) theory of Configurationality are explored. We propose to expand Baker's theory to allow "discontinuous" constituents of the type (Wh$ sb1$ ... OP$ sb1$) to account for Island effects in IHRCs of pronominal languages like Mohawk and Navajo. / No special rules for the interpretation of IHRCs are required. The LF-Raising rule proposed in Williamson (1987), which raises the internal head at LF to the COMP of the IHRC is shown to derive from the LCR as it is required to create an operator-variable (A$ sp prime$-Binding) relation. We provide further independent evidence for this rule in Chapter 2.

This thesis explores the connection between linguistic theory, as embodied in a version of the Government - Binding (GB) model of syntax, and the parameter-setting theory of language acquisition. In Chapter 2, it is argued that by incorporating the criterion of epistemological priority, syntactic theory can move closer towards becoming a plausible model of language acquisition. A version of GB theory is developed which adopts this criterion, leading to several modifications, including the derivation of X-bar theory from more "primitive" grammatical sub-components, and a revision of the Projection Principle. This model is converted into a procedure for phrase-structure acquisition, employing sets of Canonical Government Configurations and Percolation Principles to map Case- and θ-relations onto phrase-structure trees. The chapter ends with a discussion of the "missing-subject" stage in the acquisition of English. Chapter 3 concerns auxiliaries. It is argued that parametric variation in auxiliary systems can be reduced to levels of association between INFL and V. The question of irregularity is dealt with through the Designation Convention of Emonds (1985), which makes a distinction between open- and closed- class grammatical elements, and a Parallel Distributed Processing model of learning. The last part of the chapter investigates the learning of the English auxiliary system, and in particular the errors known as "auxiliary overmarking". Chapter 4 investigates the syntax of Subject Auxiliary Inversion (SAI)-type rules. An account of inversion is developed based on the theory of predication, in which inversion-inducing elements are treated as "A'-type" subjects which must be linked to AGR in order to satisfy conditions on Predicate-licensing. A parametrization is developed based on the cross-linguistic examination of SAI-type rules. Chapter 5 concerns the acquisition of SAI. It is argued that there are no invariant "stages" in the development of inversion; rather, a proportion of children misanalyze (WH + contracted auxiliary) sequences as (WH + AGR-clitic) sequences and formulate grammars in which SAI is unnecessary. A "two-tiered" theory of syntactic acquisition is proposed to account for the observed developmental patterns. / Arts, Faculty of / Linguistics, Department of / Graduate

Traditional views of grammatical theory hold that languages are characterised by sets of constructions. This approach entails the enumeration of all possible constructions for each language being described. Current theories of transformational generative grammar have established an alternative position. Specifically, Chomsky's Government-Binding theory proposes a system of principles which are common to human language. Such a theory is referred to as a "Universal Grammar"(UG). Associated with the principles of grammar are parameters of variation which account for the diversity of human languages. The grammar for a particular language is known as a "Core Grammar", and is characterised by an appropriately parametrised instance of UG. Despite these advances in linguistic theory, construction-based approaches have remained the status quo within the field of natural language processing. This thesis investigates the possibility of developing a principle-based system which reflects the modular nature of the linguistic theory. That is, rather than stipulating the possible constructions of a language, a system is developed which uses the principles of grammar and language specific parameters to parse language. Specifically, a system-is presented which performs syntactic analysis and translation for a subset of English and German. The cross-linguistic nature of the theory is reflected by the system which can be considered a procedural model of UG. / Science, Faculty of / Computer Science, Department of / Graduate