In sports discourse, the relationship between athletics and technology is often paradoxical. On the one hand, modern sports rely on technology at every level, from training and tracking of players to the equipment and apparel used by athletes to the game strategies and playing fields themselves. Nearly all of these technologies are intended to increase athletic performance on some level. And yet, certain performance enhancement technologies can be criticized for being antithetical to the spirit of sports, which is framed as being a strictly natural and pure human endeavor. Using a rhetorical-cultural methodological approach, popular sports discourse is analyzed to investigate how arguments in contested spaces between sports and technologies get (re)negotiated and (re)articulated to fit within a sports social language that emphasizes "pure" and "natural" ideals of sport. This often results in a dichotomy where the sport/technology relationship is either black boxed, thus being subsumed in the sport social language and becoming transparent and the relationships unarticulated, or the technology is regulated out of the sport through rules and bans. The reason for this articulation is attributed in large part to the deep humanism embedded in the sport social language. How a shift to a posthuman perspective would effect sports discourse is explored. These conclusions about underlying values in sports discourse lead to the formation of a new theoretical framework called disciplinary mythologies. Building off of Foucault's disciplinary power, Scott's disciplinary rhetorics, and Barthe's mythologies, disciplinary mythologies are discrete units of persuasion that both construct and constitute claims by drawing upon layered narratives and shifting associations that lose their context when entering the realm of myth. Two specific disciplinary mythologies are discussed—the level-playing-field topos and the nostalgia enthymeme—and it is shown how sports discourse often draws upon them to shape arguments and actions.

Submitted by Irene Nascimento (irene.kessia@ufpe.br) on 2016-08-04T18:54:26Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) DissertacaoFinalDominiqueDigital.pdf: 1809626 bytes, checksum: 0394869923ec4dde774f79a5ec5290de (MD5) / Made available in DSpace on 2016-08-04T18:54:26Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) DissertacaoFinalDominiqueDigital.pdf: 1809626 bytes, checksum: 0394869923ec4dde774f79a5ec5290de (MD5) Previous issue date: 2016-02-29 / Facepe / A presente pesquisa se desenvolveu no âmbito do Observatório Temático e Laboratório – Ensino, Tecnologia, Ciência e Informação (OtletCI) com a intensão de avançar na questão de como extrair informação relevante e de como representá-la para fins de recuperação semântica da informação, em particular no caso de textos de publicações científicas em português. Para tanto, como metodologia, investigou-se a tecnologia da busca semântica quanto aos fundamentos teóricos, sua utilidade no contexto do OtletCI e requisitos para aplicação em textos científicos em português. Como experimento, buscou-se explicitar os requisitos da busca semântica para a aplicação em textos científicos, através da análise da extração de relacionamentos semânticos do tipo “causa e efeito” em 60 resumos, em português, de artigos científicos da área de Ciências Agrárias. O estudo apresentou, por meio de considerações de ordem qualitativa e quantitativa, uma comparação entre o processo manual e automático de extração de sentenças de causa e efeito. Esses documentos foram previamente analisados de forma manual, e as sentenças de causa e efeito foram extraídas através da leitura dos resumos. Para o processo automático, com os dados transferidos do software PALAVRAS para a planilha do Excel, foi possível realizar uma programação para localizar sentenças de causa e efeito automaticamente. O objetivo foi comparar as sentenças identificadas diretamente pelo pesquisador e as sentenças reconstruídas automaticamente a partir do conjunto de células programadas. Conclui-se enfatizando que a possibilidade de usar técnicas automáticas acelera o processo de criação e extração de relações de causa e efeito e pode ser usada como alternativa ao processo custoso de identificação manual de informações semânticas. Porém, mais importante que propor uma estrutura de relações de causa e efeito para a construção de sistemas de busca, o que pode-se apontar como o resultado mais expressivo da presente pesquisa é o estabelecimento preliminar de rotinas para a versão automatizada. / This research is developed within the Thematic Observatory and Laboratory - Education, Technology, Science and Information (OtletCI) with the intention to move forward on the question of how to extract relevant information and how to represent it for purposes of semantic retrieval of information, particularly in the case of texts of scientific publications in Portuguese. Therefore, as a methodology, we investigated the semantic search technology based on the theoretical foundations, its usefulness in the context of OtletCI and requirements for application in scientific texts in Portuguese. As an experiment, we tried to clarify the semantic search requirements for the application of scientific texts by analyzing the extraction of semantic relationships such as "cause and effect" in 60 abstracts, in Portuguese, of scientific articles in the area of Agricultural Sciences. The study shows, through qualitative and quantitative considerations, a comparison between manual and automatic extraction process of cause and effect sentences. These documents were previously analyzed manually, and the sentences of cause and effect were extracted by reading the summaries. For automatic process, with data transferred from PALAVRAS software to the Excel spreadsheet, it was possible to carry out a program to find cause and effect sentences automatically. The goal was to buy the sentences identified directly by the researcher and sentences automatically reconstructed from the set of programmed cells. The research concludes emphasizing that the possibility of using automatic techniques accelerates the process of creating and extracting of cause and effect relationship and may be used as an alternative to costly manual process of identifying semantic information. However, more important than to propose a structure of cause and effect relationships for building search engines, we can point out as the most significant result of this research the preliminary establishment of routines for automated version.