M.A. / This study is based on the principles of the subjectivist approach of Cognitive Linguistics, as opposed to the more traditional objectivist view, specifically those principles applicable to the acquisition of mental contact between conceptualizers in a given communication situation. One of the most fundamental points of departure of Cognitive Linguistics is the opinion that abstractions (even linguistic abstractions) are modelled on man's bodily experience of his surrounding reality. These embodied experiences constitute a network of preconceptual and non-propositional image schemas, categorized as space schemas, force dynamic schemas, schemas based on sensation and basic-level objects. Several mapping processes, including metaphor and metonymy, transpose these image schemas from a preconceptual, prelinguistic level to a conceptual and linguistic niveau. A prerequisite for conceptualization through linguistic communication is to constitute mental contact between the speaker and hearer as the conceptualizers. Linguistic communication presupposes the transfer of meaning, which is based on certain cognitive variables determining mental spaces and conceptual blends. Against the preceding background, the newspaper headline, as cognitive entity, constitutes the research domain by means of which the nature of the potential mental contact between the headline writer (as speaker) and the headline reader (as hearer) in a specific communication situation is analysed and evaluated. A seemingly useful and potent cognitive measuring instrument regarding presupposed and actualised mental contact, by means of which any written or spoken communication can be analysed and evaluated, is the outcome of this study. A set of examples of headlines was collected from the Afrikaans daily newspaper, Beeld, and the Afrikaans Sunday paper, Rapport, from February 1996 until September 1998, serving as research material.

M.A. (Afrikaans en Nederlands) / Please refer to full text to view abstract

This paper is about the sound iconicity of the Late Middle Chinese entering tone in two of Du Fu's long narrative poems, "The Journey to the North" and "Singing my Heart out in Five Hundred Characters on the Way from the Capital to Fongxian County", as well as Du Fu's grammar of poetry in these two poems. In poetry, rhyme is an arbitrary and 'visible' figure reiterated with regulation which forms an axis of sequence, and this axis will work jointly with all the other poetic elements—semantics, images, and grammar to form the whole of a poem. In these two poems, all the rhyme characters carry a voiceless —t ending, which is classified with —k and —p endings as the entering tone in Late Middle Chinese reconstructed by Edwin Pulleyblank. These voiceless stops are short, tense, and uncomfortable to utter; when they are repeated fifty and seventy times at the end of each couplet, it naturally brings about a strong, rough, and uncomfortable feeling which correlates with the feeling of suffering in both poems. It is sound iconicity, because an icon resembles the object it stands for in an immediate and concrete manner, and the —t ending rhyme characters do have the characteristics to make the reader grasp the feeling of suffering when she reads the poems. In terms of Du Fu's grammar of poetry, I used Jakobsonian methodology and found how Du Fu's poeticity was created with lexical meaning and grammar. Although Classical Chinese does not have a huge grammatical repertoire (e.g., person, case, gender, finite, non-finite . . .) which can figure in a poem, this language still has its own obligatory categories that will provide for the 'grammar of poetry'. Classical Chinese is already known for its grammatical parallelism in poetry, because this language is extremely isolating and analytical. However, grammatical parallelism is little in these two poems, but there are different kinds of grammatical tropes. They are mainly anti-syntactic inversions interacting with semantics. I found Du Fu a fascinating artist of grammar; he may be anti-grammatical but never agrammatical. / Arts, Faculty of / Asian Studies, Department of / Graduate

The structure of our world is given in the grammar of our native tongue. If so people whose native tongue has quite a different grammar must be living in a quite different world. A logic such as Aristotle's may seem universal to the speakers of Greek, in fact it may seem universal to speakers of any Indo-European tongue, but the logic will hold good only for the 'universe' of the language or language-family in question. This implies a relationship between logic and grammar rather like the one Russell and Whitehead -claimed for mathematics and logic. Their Principia Mathematica tried to show that the mathematical notion of number rests on, or arises out of, the logical notion of class, that is, we come to understand what a number is through our grasp of what a class is. This thesis is a kind of Principia Logica: it suggests that the whole framework of common sense logic rests on, or arises out of, the grammatical structure of the language the logic was conceived in or took shape in terms of. And if so logical criteria come into being and take shape inside a language or language-family, and are dependent for their validity and even for their meaning on the structure of the language in question. To test, or to try to test, a mode of thought or an argument form against a logical system would be to put the cart before the horse: the logic only makes sense because the form of argument or mode of thought was there already. If so philosophers and logicians ought to think of the words 'world', 'universe' and 'universal' with tongue in cheek. In so far as a judgment seems to us universally true it is unlikely to hold good for the world of an alien language family. If our world is not the only world anybody writing logic or philosophy down ought to make it clear whose world he has in mind - and to do this it may be enough to make sure it is addressed to somebody in particular. Western philosophers seem to have addressed themselves to the whole world, or to mankind, or God. This thesis shows, if nothing else, how hard it can be to address even one other human being. To sum up with another analogy; it seems to me, as a single man, that the difference between one and two is greater than the difference between any other two numbers. There may be a world of difference between zero and one, but between one and two there's all the difference in the world - and that's the difference that matters. Perhaps the only way 1,000 differs from 1,001, as Frege puts it, is in the expression on its face. / Arts, Faculty of / Philosophy, Department of / Graduate

Problem: The purpose in undertaking this study was to determine more exactly the contemporary usage of the Spanish relative pronouns, thus making it possible (1) to supplement existing descriptions of the usage of the relative pronouns, and (2) to restate contemporary usage. Method: For this study, representative works of the best contemporary Spanish and Spanish-American authors were used, employing Keniston's sampling technique. Three passages of ten pages each were selected from the beginning, the middle, and the end of the work in question. All examples found were recorded individually for analysis, thus making possible a study of a much larger number of actual examples than are available from the Keniston study. Conclusion: A. The following findings which have been established in this thesis in some cases supplement existing descriptions by the authorities and in others, are at variance with them. 1. Que is not frequently replaced after con by el cual to avoid confusion with conque. 2. Que does not necessarily follow its antecedent immediately. In fifteen per cent of all cases observed que was removed from its antecedent in such a way that determination of its antecedent was difficult. 3. Que does on rare occasions as object of a preposition refer to a person. 4. Quien is frequently used to distinguish a person from a thing, as antecedent. 5. Quien as a substantive refers more frequently to indefinite persons than el que. 6. Quien as a substantive is used almost exclusively after como (meaning "as one who"). 7. El que, as well as el cual, is regularly removed from its antecedent by a comma (or similar punctuation, and/or one or more words.) B. Restatement of Contemporary Usage. 1. In a restrictive clause: a. The subject is regularly que for persons and things. b. The object is usually que for persons and things. Quien is used with "personal a". Al cual and al que are rarely so used. 2. In a non-restrictive clause: a. The subject is usually que for persons and things, but may be quien, el cual, and to a lesser extent, el que. b. The object is usually que for persons and things; quien is used frequently with "personal a"; al cual or al que, rarely. 3. Object of a (exclusive of "personal a"), en, de and con: a. Referring to persons, quien, el cual or el que are used. In non-restrictive clauses quien is most frequently used. b. Referring to things, que, el cual or el que are used. Que is normal in restrictive clauses; que, el cual or el que in non-restrictive. 4. Object of por, sin and tras: a. El cual or el que are used; possibly, quien. 5. Object of dissyllabic prepositions: a. El cual or el que are normally used. 6. Object of compound preposition: a. El cual is normally used; el que, rarely. NOTE: Where there are choices in the usage indicated throughout Section B to this point, variety, rhythm and euphony are considerations. However, the choice of quien, el cual and el que to avoid ambiguity must be particularly borne in mind. 7. Lo cual and lo que as restrictive and non-restrictive relatives: a. After monosyllabic prepositions both are used. b. After dissyllabic and compound prepositions, only lo cual is used. 8. Quien and el que as substantive relatives: a. As object of a preposition, quien and el que are both used, the latter more frequently. b. Referring to indefinite persons, quien is used more frequently. c. Referring to definite persons, el que is used more frequently. d. After como (meaning "as one who") quien is used. / Arts, Faculty of / French, Hispanic, and Italian Studies, Department of / Graduate

This thesis presents a linguistic description of Gurma nouns and verbs based on the Fada N'Gurma dialect. The first chapter introduces the Gurmas to the reader and tells about the first writings in and about their language. Chapter two presents a phonetic and phonemic analysis of the language. Chapter three defines the noun, presents the noun classes, and shows how tonal patterns divide the noun classes into sub-sets. It shows the pronouns which are in concord with the noun classes as well as the sub-set of personal pronouns. It then shows how tonal patterns change when nouns or pronouns are used in negative, relative and possessive constructions. Chapter four defines the verb and shows its inflections by means of tonal change, addition, deletion, replacement and suppletion. It then presents the verbal particles and shows their use. In chapter five a Gurma sentence is analysed to shown nominal and verbal morphemes in an authentic Gurma setting. The thesis is concluded with a bibliography and two appendixes, the first a brief report on work being done in the currently popular field of ideophones, the second an account of references made to the Gurma language before the twentieth century. The two maps following, page ii, show the location of Gurma in relation to Greenberg's language families of Africa, and in relation to Gurma's immediate language neighbours. / Arts, Faculty of / Linguistics, Department of / Graduate

This study reports on some of the features and practices of sense making involved in the work of explicating the normative order of language use. The normative order of interest is (what Wittgenstein would call) the grammar of the concept justification. The data consists in the author's mundane work of figuring out in what context some type of talk would count as "doing justifying." The warrant for this enterprise issues from ethnomethodology's concern with sense making, or practical reasoning. While ethnomethodologists have addressed accounting and interpretive practices, the phenomenon under investigation here is of a different, previously unexplored type: "figuring out." In order to prepare for the analysis consideration is given in Part One to three questions: What are the "features and practices of sense making?" What analytic status must such features and practices have in order to be of ethnomethodological interest? How can the features and practices of figuring out grammar be best studied? Ordinary language philosophy is drawn on to answer the first two questions. A distinction is made between natural and social science in terms of the source of, and constraints on their concept formation and use. This distinction provides for seeing why it is that natural science can have a technical language while social science only can make technical use of ordinary language. That technical use is argued to depend for its sensibility cm the indexical limits of ordinary use. Features and practices of sense making thus turn out to be whatever members sanctionably can call features and practices of sense making. Some claims in ethnomethodology are found not to meet this indexical criterion. The answer to the second question has been that invariant or formal properties are of interest. Different versions of invariance are located in the literature. Using Wittgenstein's argument against essentialism the search for universal invariance is rendered questionable. Instead, particular invariance and type-invariance are claimed to be discoverable and warrantable within the limits of ordinary language. In addition the argument is put forward that (repeatable) contingent practices deserve attention. The third question is answered by considering and comparing a third person and a first person approach to studying figuring out. In terms of contingent practices it is found that a third person approach faces a problem of indexicality, whereas a first person approach does not, or if it does, it can survive it. Methods of ethnomethodological reduction and eidetic variation are discussed. In Part Two the concepts of meaning, force and grammar are introduced and explicated. These concepts are then used in presenting the normative order that governs the use of the concept of justification. Consideration is then given to how the generation of that normative order (grammar) is to be viewed analytically. Using a first person approach in Part Three, the author's own work of figuring out the grammar of justification becomes the topic of study. That study is written and furnished as a "journey": the analysis of each practice is developed in response to the properties of the phenomena, and each analysis draws and builds on the prior one. Four contingent practices are analyzed: calling, filling, grounding, and answering-seeking. As well, three features are found to be essential to figuring out: pre-reflective knowing, pre-reflective awareness of possibly knowing and orienting-to-grammaticality. Together these seven properties reveal that figuring out has a structure fundamentally different from accounting and interpreting. Through a consideration of these seven properties and other features an argument is provided in Part Four against using the Weber-Schutz version of social action as a resource for defining ethnomethodology's domain. Instead, it is argued that ethnomethodology's domain and phenomena are coextensive: the social is sense making. / Arts, Faculty of / Sociology, Department of / Graduate

This thesis provides an analysis of Possessor Raising (PR) in a minimalist framework. I examine four languages that exhibit PR, namely Japanese, Korean, Kinyarwanda, and Swahili. I propose that cross-linguistic variation of PR in these languages is captured by the single notion of Multiple Feature-Checking (MFC). In addition to cross-linguistic variation of surface syntax of PR, this thesis also examines a universal feature of PR, namely the relational noun restriction. PR raises an interesting problem with the mapping relation between an argument DP and its grammatical function (GF). A DP is normally associated with a unique GF (i.e., a one-to-one mapping relation), but in some cases such as passive, a DP may be associated with more than one GF, being both an underlying object and a surface subject (i.e., a one-to-many). PR also poses another type of mapping relation, namely a many-to-one relation; under PR, a possessor DP may also bear the GF of its host. In order to capture such a many-to-one relation between a DP and its GF in PR, I propose that this is an example of MFC, which is defined as follows: (i) MFC is possible only if a Formal Feature (FF) of a head (T and v) can escape deletion. (ii) MFC applies to both strong and weak FFs. In contrast to the original formulation of MFC (cf. Chomsky 1995b), which always involves overt movement of DPs and derives multiple specifiers, I argue that MFC also takes place at LF, which involves covert movement of FFs and derives multiple adjuncts of feature bundles onto a head. Given the evidence against overt movement of the subject and object in Japanese and Korean, I argue that PR in these languages is best analyzed in terms of covert MFC. I also provide an overt MFC analysis for Kinyarwanda and Swahili PR. Lastly, I discuss the relational noun restriction on PR, which holds of all PR languages. I propose that this restriction is reducible to the structural position of the possessor of a relational noun, namely its position as complement to the noun. / Arts, Faculty of / Linguistics, Department of / Graduate

As the title indicates, this study COHESION IN A TSWANA NARRATIVE TEXT aims at examining the occurrence of cohesion in a selected Tswana text. The study comprises eight chapters. The introductory chapter outlines the aim of the study, the method of research, the division of chapters, and further provides an explanation of key concepts, namely cohesion and coherence. Cohesion refers to the flowing together of ideas in a text thus adding to its coherence. Coherence is referred to as a mental phenomenon, something the reader establishes or wishes to establish. The five different types of cohesion as classified by Halliday (1994) (for English) are reference, ellipsis, substitution, conjunction and lexical cohesion. Each of these cohesive ties has different categories. The aim is to establish how applicable they are to Tswana. These ties operate in different ways i.e. anaphorically, cataphorically and/or exophorically. These forms are concerned with the direction in the text. Ties that operate anaphorically point backwards to what precedes the text; those that operate cataphorically point to what is to follow and ties that operate exophorically point to that which is outside the text. They also reflect on the distance in terms of number of sentences or clauses between the cohesive item and the item it refers to. The examples are taken from Tswana text Dilo di Masoke by Monyaise (1991). The general findings and overview of the research are outlined in the concluding chapter. / Dr. J.H.M. Kock

Standard Thai classifiers have never been studied exclusively and comprehensively. That is, they have been included in grammars of the entire language and are usually discussed by giving a few examples. Specific papers usally deal with some particular aspect of classifiers exclusively. The criteria of classification by a given classifier are a puzzle in many cases. Often there are obvious semantic criteria of classification, and obvious syntactic criteria as well. The relation between semantic and syntactic criteria of classification is unclear. Equally unclear is the relation between headword and classifier, a relation which is the basis of the criteria of classification. In this study, a near-complete list of Thai classifiers is drawn from the Thai-English Student's Dictionary by M.R. Haas (1965). A semantic definition of classifiers is provided and various categories are described. The first category is Repeaters. Members of this category are found to be one-place predicates. A second category is Partial Repeaters and these are found to have a relationship of hyponymy with the headword. Measures are another important category, divided into Standard Measures (exact non-entities) and Temporary Measures (inexact units used as measures according to the intention of the speaker). The remaining category is found to contain non-compoundable classifiers of two types: those with very general referents (size, shape, etc.) and those classifiers in various stages of decreasing semantic equivalence with the sense of the same form as full substantive noun. The former type is labelled "General Units", the latter "Extended classifiers". Extended Classifiers are seen to be of two types: those with a single sense (or apparently "meaningless") and those with several senses, only some of which qualify as Extended Classifiers. The apparently meaningless classifiers /lem/ and /an/ do not occur as substantive nouns at all. The complete collection of nouns classified by each of these two classifiers (according to Haas 1965) is made and each class is examined for common semantic features. On the basis of comparative information from neighbouring and related languages, various possible criteria of extension are established for /lem/, and a basic sense equivalent to "sickle" is indicated by the data. The case with /an/ is rather different since its extensions are much wider and a larger number of nouns is classified. Based on prominent compounds classified by /an/, a basic sense of "stick" is hypothesized for /an/ and the possible paths of semantic extension are suggested. Finally, a "meaningful" Extended Classifier (/tua/, "body") is examined. The sense attributed to this form by native speakers when it functions as a full substantive noun becomes a factor in choosing the sense "body-shape" as the basic sense with classifier function. In concluding remarks, suggestions are made about the applications of the method and results of this paper to lexi cography and semantic reconstruction in Thai and languages of the area. The necessity of further phonological, histori cal and ethnographic information is stressed. / Arts, Faculty of / Linguistics, Department of / Graduate