This is the first in-depth, quantitative study of register variation in Scottish Gaelic (ScG), an endangered Celtic language. Previous work on the subject has been mainly anecdotal in nature or a by-product of other lines of investigation. A recent diachronic study on Gaelic ‘news-speak’ (Lamb 1999) found it impossible to fully uncover the characteristics of the register without benchmark data on the language’s overall register range and morphosyntactic variation. This provided the impetus for the current study, which compares eight varieties of spoken and written ScG. An 81,000 word computerised corpus of Gaelic texts was compiled and tagged for a wide range of linguistic features sensitive to register differences. The tagging process was informed by a new descriptive grammar of ScG, which is included as an appendix. The registers were compared on the basis of the distributions of these features utilising several inferential statistical tests (e.g. Analysis of Variance). The study focuses on three broad areas of linguistic variation: syntax and information structure; morphology; and NP grammar and complexity. Robust differences were found between most of the register types in the corpus. In particular, conversational ScG stood out as markedly different; e.g. it had simpler NPs, fewer modifiers, less ‘subordinate’ structures and a greater use of fragmented structures such as left-detachment and asyndetic clausal juxtaposition. These features are similar to the attributes of spontaneous spoken language described by Chafe (1982) and Miller and Weinert (1998). Other major contrasts obtained for narrative vs. non-narrative, reportage vs. non-reportage, and formal writing vs. other register types. Five underlying contextual parameters were found to be responsible for the variation present; 1) production constraints; 2) discourse freedom; 3) information orientation; 4) interaction; and 5) producer characteristics. Overall, the results correlate well with those of Biber (1988 <i>inter alia</i>) and others adopting multi-dimensional models. They also demonstrate that ScG, despite being an endangered language, has a level of register variation comparable to that found in larger languages such as English.

This Thesis reflects my efforts to build, up a theoret¬ ical framework for the analysis of discursive capacity. By discursive capacity is meant the various procedures used by competent readers for the interpretation of technical and scientific English discourse. A distinction is made between "rules" and "procedures". Among the rules that are consider¬ ed in the first part of the thesis are those relating to "grammatical", "textual", and "sociolingulstic" competences. These competences are associated with meanings referred to, respectively, as "literal", "situational", and "contextual" or "pragmatic". The sociolinguitic approach to discourse is surveyed and also, to some extent, criticised. In the second part of the thesis this approach iB modified by taking account of some notions discussed, particularly, by certain philosophers of language. These notions are related to discourse analysis; they included distinctions between, for instance; text and textualisation, process of interpretation and product of inter¬ pretation, the primary and secondary system of a discourse, statement and discourse-act. The theoretical discussion is interspersed with analyses of specimen texts. On the basis of the theoretical discussion there are put forward a number of propositions embodying certain pedagogic implications of the 'theory1. In the third part of the thesis these pedagogic implications are explained and illustrated. A number of exercise prototypes are then suggested.

No description available.

This thesis is a diacronic and synchronic study of the writing system of the standard Thai language. It is divided into three parts. Part I, which is the introductory part, contains two chapters. Chapter One discusses the relationship between speech and writing and a typology of writing systems. A brief sketch of the characteristics, the development and the phonological system of the Thai language is presented in Chapter Two. Part II, concerning my original investigation on the ancient Thai writing system, is composed of three chapters. Chapter Three details characteristics of the first and the second oldest known Thai scripts called the Sukhothaiscript (1293 A.D.) and King Li Thai script (1357 A.D.). The question of the origin of the Thai script from the cursive form of the ancient Cambodian script is considered in the light of cultural, archaeological, epigraphic and linguistic evidence. The phonetic interpretation of proto-Thai sounds on the basis of historical spelling is also discussed in this chapter. Chapter Four analyses the phonetic correlates of the orthography and the characteristics of the third oldest known Thai script, King Narai script (1680 A.D.). The gradual development of the Thai sound symbols from the Sukhothai script, King Li Thai script and King Narai script is shown in detail. Chapter Five describes the proposed reform of the Thai writing in 1917 by King Rame VI. The invention of new characters and setting up new rules of orthography which the king suggested for the Thai writing system are also presented. Part III, which contains seven chapters, is my own phonetic analysis of the relationship, or the 'fit', between the sounds of the present Thai language and the symbols that are used to represent them. Discussion on the closeness of the fit between sounds and symbols, i.e. in the regularity with which a given symbol represents a given sound and the problems in reading and writing the script, is presented in various places throughout this part. Chapter Six describes consonant symbols. Chapter Seven discusses syllable-initial consonant symbols. Vocalic symbols are reviewed in Chapter Eight. Chapter Nine is concerned with syllable final consonant symbols. Chapter Ten deals with tonal markers. Chapter Eleven concerns punctuation, mathematical symbols, distinctive treatment of words, spaces, numerals and abbreviations. Writing styles are illustrated in the final chapter, Chapter Twelve. The five appendices include a brief survey of earlier attempts to study the Thai script since 1854 A.D., a poem containing the designatory names of the forty-four Thai consonant symbols, the general system of transliteration of Thai characters into Roman invented by the Royal Thai Academy, notes on the transliteration of English words into Thai characters, and suggestions using Thai phonetic symbols in transcribing Thai dialects.

No description available.

This work is a study of the vocabulary of colour in modern standard French carried out within a framework of structural semantics. The first chapter contains a general discussion of semantic structure with particular reference to the relativity hypothesis and semantic universals. Then follows an analysis of data from three sources (1) colour terms elicited directly from French native speakers (2) colour terms elicited using the questionnaire method and (3) colour terms from the computer corpus of the Tresor de la Longue Francaise. The notion of basic colour terms is examined and the correlation between basicness and frequency is investigated. An account is given of experiments carried out by French speakers in order to establish (a) the best examples (foci) of colours designated by basic terms and (b) the boundaries netween them. Results of these experiments are compared with the results of similar experiments carried out on other languages and the degree of cross-cultural agreement about colour foci and colour boundaries is ascertained. Colour terms are considered not only in the language as a whole but also in sub-systems of the vocabulary and in context. Collocationally restricted terms are examined and in particular the uses of 'brun' and 'murron' are investigated. Physical and physiological aspects of colour are examined , particularly with reference to the notion of lexical opposition. Basic colour adjectives are examined on the levels of inflectional and derivational morphology, special attention being given to approximate colour terms. The concluding chapters contain a discussion of the key works of the past three decades by linguists, anthropologists and psychologists in the field of colour, and an account of some of the implications of the author's findings for structural semantics.

The aim of this work is two-fold, being, firstly, to describe certain important areas of the syntax of modern spoken Telugu and, secondly, by formulating the description in terms of localise case theory, to contribute something to the understanding of this approach. Illustrative support for the localist hypothesis has hitherto been drawn largely from languages of the Indo-European family. This thesis attempts to provide a supplement to this by drawing on a language of the Dravidian family, which is genetically entirely unrelated to Indo-European. After an introductory chapter surveying earlier work on the grammar of Telugu, from which it will become apparent that there has so far been only a modest amount of work on Telugu syntax in terms of any recent theoretical model and nothing in terms of the localist approach, there follows an account of Telugu sentence patterns and the major elements of Telugu sentence construction. This is intended to provide the necessary data on the surface grammar of the language to make the subsequent proposals regarding underlying structures more readily comprehensible. Chapter 3 gives an outline of the theory of generative localism and makes comparisons with other current syntactic theories, in particular the case grammar of Fillmore and generative semantics. The fourth chapter examines one of the primitive case relations, namely the one that specifically underlies the spatial locative constructions of Telugu. Attention is drawn in the discussion to the co-relationship between word-order and definiteness and the bearing this has on a statement of the derivation of existential clauses. Chapter 5 attempts to show the underlying uniformity between the concrete locatives (spatial locatives) and certain apparently nonconcrete phenomena such as possessive and stative expressions, including verbs of cognition, perception and wanting. The examination of equative clauses that is presented'-in Chapter & focuses attention on agu, the 'copula' of Telugu. The presentation leads to the conclusion that this 'copula' needs to be taken as a basic verb rather than as a dummy element. It also becomes necessary to suggest the need to abandon the Fillmorean principle of one instance per clause of a given case relation. The locative discussed in Chapters 4 and 5 has been non-directional. Chapter 7 is concerned with directional locatives, that is to say with 'source' and 'goal'. It is shown that these two do not involve two further case relations, since 'goal' can be accounted for as a sub¬ type of static locative. The thesis concludes with a summary account of further worthwhile avenues of research in Telugu syntax which might usefully be handled in localist terms.

No description available.

The present work attempts to establish a theory of reference from a linguistic - rather than philosophical - point of view. PART Is The Preliminaries (pp. lo-113) surveys various linguistic and philosophical problems associated with ref¬ erence; it argues against predicational analysis as a vi¬ able framework for dealing with reference; and it estab¬ lishes the field of referentiality as the domain divided between deixis (spatio-temporal location) and denotation (categorial location). PART II: The Theory (pp. 114-233) begins by drawing a fundamental distinction (based on Frege) between syntactico- semantic (SS) and referential-semantic (RS) analysis, and by setting up the notion "referential potential" as a property of linguistic items. The common — metaphysical — basis for the calculus of classes and the referential theory is demonstrated, and the formal framework developed. Four RS-categories are recognized, associated with "all", "kind" (i.e. genus), "some", and "one". These categories are considered to be the 'heads' in referential phrases, each one of which consists of one of the functional (deictic) categories and one lexical (denotative) category. The referential phrases are ordered hierarchically in a referential branch under which NP is generated. Two different serialization-types (appositive and delimitative) are considered referentially significant. A transformational component is introduced to account for serialization within the NP; three transformational pro¬ cesses are recognized. After a number of data from languages other than Eng¬ lish has been adduced in support of various aspects of the theory, PART IIIi The Application (pp. 234-413) be¬ gins by establishing the (closed) class of referential functives in English. These fall into four subclasses: quantifiers, E-classifiers, determinatives, and pronouns. These subclasses are established distributionally on the basis of the serialization-types they may enter. The remainder of PART III applies the theory to English NP's which contain a referential functive. Finally, a number of other areas are briefly mentioned for which RS-analysis is likely to prove insightful.

No description available.