This thesis investigates the phenomenon of polysemy: a single lexical form with two or multiple related senses (e.g. catch the rabbit/order the rabbit; lose a wallet/lose a relative; a handsome man/a handsome gift). I develop a pragmatic account of polysemy within the framework of Sperber and Wilson’s relevance theory, where new senses for a word are constructed during on-line comprehension by means of a single process of ad hoc concept construction, which adjusts the meanings of individual words in different directions. While polysemy is largely unproblematic from the perspective of communication, it poses a range of theoretical and descriptive problems. This is sometimes termed the polysemy paradox. A widely held view in lexical semantics is that word meanings must consist of complex representations in order to capture the sense relations involved in polysemy. Contrary to this view, I argue that a conceptual atomist approach, which treats word meanings as unstructured atoms and thereby avoids the range of problems associated with decompositional theories of word meaning, may be at least as able to account for polysemy when paired with an adequate pragmatic theory. My proposed solution to the polysemy paradox is to treat polysemy as a fundamentally communicative phenomenon, which arises as a result of encoded lexical concepts being massively underdetermining of speaker-intended concepts, and is grounded in our pragmatic inferential ability. According to this approach, the role of the linguistic system in giving rise to polysemy is to provide a minimal input, or clue, which the pragmatic system uses as evidence to yield hypotheses about occasion-specific, speaker-intended meanings. I further show how this pragmatic approach can account for cases of ‘systematic polysemy’, usually seen as prime candidates for an analysis in terms of lexical rule application. Finally, I develop an account of metonymy within the overall framework of relevance-theory.

This thesis outlines a model of Speaker-Addressee interaction that suggests some answers to two linked problems current in speech production. The first concerns an under-researched issue in psycholinguistics: how are decisions about speech content – conceptualization – carried out? The second, a pragmatics problem, asks how Speakers, working under the heavy time pressures of normal dialogue, achieve optimal relevance often enough for successful communication to take place. Links between these problems are discussed in Chapter 1; Chapter 2 reviews existing research on speech production and dialogue. Chapter 3 presents the central claim of my thesis: that the Addressee exerts a significant influence over the Speaker’s decision-making at a level below the latter’s consciousness. Using evidence drawn from psycholinguistics, developmental psychology and human-computer interaction, Chapter 4 presents evidence to support this claim, demonstrating that a Speaker’s performance can be decisively affected at a preconscious level by the degree of attentiveness shown by the Addressee. Lack of attentiveness, in particular, appears to damage speech production at the conceptualization level. I suggest, therefore, that Speaker and Addressee are linked in a feedback loop: unless a Speaker achieves and maintains relevance to an Addressee, the Addressee’s interest will be lost, and this will impair the Speaker’s production abilities and hence the communication process itself. Chapters 5 and 6 consider some automatic mechanisms that may help Speakers dovetail their productions to Addressee need. These include the neural mechanisms underlying face perception and social rejection; automatic aspects of theory of mind; intuitive memory and inference systems of the type being explored in dual-process theory; and connections between verbal performance and behavioural priming currently being investigated. Chapter 7 summarizes the complete argument, discusses its wider implications, and includes suggestions for further work.

This thesis examines the vitality of Hidalgo Nahuatl (HN) in the communities of Jaltocan, Panacaxtlan, Santa Cruz, Santa Teresa and Zohuala in the Huasteca Hidalguense, Mexico. The research, conducted in Mexico and St. Andrews University from 1976-1982, applies an analysis of HN within the framework of the Sociology of Language and Dependency Theory, thereby using a multi-disciplinary approach. Through an investigation of the historical, social, cultural and economic factors related to HN, the latter is embedded in its reality. HN is shown to be originally a language of dependency and oppression, supported by a long mestizo tradition of "caciquismo". It is demonstrated that an increasing number of Spanish (S) monolinguals, together with other socio-economic factors, is encouraging Nahuas to bilingualize and S:: =A. is fast becoming the new language of dependency. The Hidalgo Nahuas possess practical reasons for the acquisition of S., these being to solve their daily problems - especially land tenancy -, to communicate with the mestizo out-group and to undertake trading with non-HN speakers. However, the Nahuas are not surrendering their native language as they bilingualize, but rather, tend to limit its usage to native Nahua contexts and speakers. HN has become important to the Nahuas in order to demonstrate their ethnic identity and territoriality. The introduction of government projects to the communities, such as the Castellanizacion project or bilingual-bicultural education, are shown to be theoretically bilingual in approach, but fail to take into account sufficiently the regional Indian language in the praxis. The stable maintenance of HN is highlighted by statistical results from the word-count of recorded texts, documents and publications and the range of morphological phenomena affecting S. words in HN is described with examples from the Corpus. The linguistic interference from S. in HN is located within Dependency Theory and this author suggests the use of the term dependency word rather than loan word and dependency language, thus implying a diachronic sociological process which is reflected in HN. Extended Texts are offered as evidence of the linguistic standard of HN and attitudes of Nahuas towards their language are presented. The final conclusion is that modern HN is a viable, vital and functional language at the time of undertaking this research and demonstrates a frequent usage by a large number of speakers. HN has still not entered into:. -avital process of language death, as is the case in other Nahuatl-speaking regions of Mexico, and is still being maintained, particularly at community level, by adults and children alike.

This is a study of the linguistic situation of contemporary Bolivia carried out between 1990 and 1993. It attempts to delimit a particular speech community (that of bilingual rural school teachers in the Quechua speaking region). It started as a study for delimiting the Spanish dialects spoken in Bolivia, seeking explanations for possible deviations from standard Spanish in the influence and actions of the mother tongue, Quechua. However, as the analysis progressed, I found increasingly a certain systematicity in the characteristics of the presumed Spanish dialect. Although there existed a determined structural transference, this did not reflect merely a direct transcription from the mother tongue Quechua, since it was not always possible to determine whether it was the result of transference from this language, or if it could be explained in terms of the non-native language. Finding some analogy with the conclusions of Labov concerning the English spoken by blacks in New York, I considered that the best explaination would be to interpret the speech in question as the expression of a distinct code. In summary, this thesis comes down specifically to the demonstration, by means of the analysis of the characteristic structures of the Spanish spoken by rural school teachers in the Quechua speaking areas of Bolivia, that the code they use as their habitual medium of communication is an interlanguage in the process of forming itself into a new code of the creole type, what we call a semilanguage. The existence of the semilanguage could also be proved in the observation of a series of social and psycological factors which affect its speakers. We could see that the teachers form an intermediate group, which is the product of a process of adaptation, and in which the confluence of certain values and attitudes has provoked the rise of hybrid values and behaviour, tending to create a new order which involves a new culture and thus a new code of expression.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.