A conversão sob a luz da linguistica de corpus

• Autor(a) principal: Costa, Roberto Diniz
• Data de Publicação: 2006
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da PUC_SP
• Texto Completo: https://tede2.pucsp.br/handle/handle/13854
• topic: Conversão; Palavras convertidas na língua inglesa; Linguistica -- Processamento de dados; Conversion; Word classes converted in the English language; CNPQ::LINGUISTICA, LETRAS E ARTES::LINGUISTICA::LINGUISTICA APLICADA

The main purpose of this study was to outline, describe what is widely called conversion and show, in an empirical way, how frequent this process is in order to point out which word classes are more converted in the English language. In order to achieve this objective, this study has found support in Corpus Linguistics, which is a Linguistics field that is concerned about analyzing any language empirically through corpora, which can be defined as authentic text compilations stored in computer files. Besides being supported by Corpus Linguistics, this dissertation has focused on the study of conversion, which is also called improper derivation in the prescriptive grammar. There is a great range of studies on the conversion process; however, very few of them has demonstrated empirically how frequent conversion is and they have seldom used authentic texts in their analysis. Therefore, this study has tried to fill this gap by being able to show empirically the limits towards conversion through the use of an authentic corpus, the British National Corpus. In order to achieve the objective of this study, the following questions have been stated: 1. What are the word classes that are most and least likely to be converted? 2. What is the conversion rate, in the selected corpus, between the words that are and are not involved in the conversion process? 3. In the selected corpus, what words show the widest array of grammar classes that can possibly be converted? 4. Are there any differences concerning the occurrence of conversion VII between the most frequent words and the words that show the widest array of grammar classes that can possibly be converted? If so, why? The corpus employed in the study was the British National Corpus, compiled from 100 million words. These words have been obtained from a wide array of sources, gathered in a way they could show a wide sample of spoken and written British English. It is obvious that, by dealing with a sample of the target language, the corpus might not represent a hundred percent of the language used. This way, it will always be an incomplete sample; therefore, when analyzing corpora, one should work with the concept of probability, which is really helpful to this study, once it made possible to show how probable conversion is. The results have indicated that conversion is a frequent process in the English language, once it involves 10 word classes and more than 36 million words. Moreover, it might be suggested that conversion is a living process in the English language and, through an empirical way, it is possible to point out that verbs, prepositions and pronouns are the most intrinsically related to conversion word classes. To sum up, this dissertation hopes that it has contributed to the study of conversion, since there are few studies that have demonstrated empirically how frequent conversion is. This work also presents and discusses the research limitations and its possible applications in the future, apart from pedagogical applications of the results