Sobre a ausência de concordância nominal no português falado em Pedro Leopoldo-MG: uma abordagem variacionista

• Ano de defesa: 2010
• Autor(a) principal: Lilia Soares Miranda Santos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Sintagma nominal; Variável linguística; Pedro Leopoldo; Língua portuguesa Variação Pedro Leopoldo (MG); Lingüística; Língua portuguesa Concordância; Língua portuguesa Português falado Pedro Leopoldo (MG); Língua portuguesa Regionalismos Pedro Leopoldo (MG); Sociolingüística
• Link de acesso: http://hdl.handle.net/1843/LETR-8TEEAU

In this study we analyze the linguistic variable constituted by the absence and presence of agreement between the inflectional elements of the noun phrase (NP) in the speech of Pedro Leopoldo / Minas Gerais, Brazil, based on theoretical and methodological assumptions of the Theory of Language Variation and Change (LABOV, WEINREICH, HERZOG, 1968; LABOV, 1972, l994), and in studies by Poplack (1980), who analyzed the Puerto Rican Spanish; GUY (1981), who analyzed the Puerto Rican Spanish and the Brazilian Portuguese (BP), Braga (1977) and Scherre (1988, 1996), who examined concord in BP. Based on these studies, we assume that the variable taken as object of linguistic analysis here is conditioned by linguistic groups of factors (NP core element: location; NP core element: grammatical class; NP non-nuclear: position; NP non-nuclear element: grammatical class; non-nuclear part of the NP: presence and absence of plural inflexion), and also by extralinguistic factors (gender, age, education and social group). The analysis is guided by these hypotheses: (i) inhabitants of Pedro Leopoldo use more often the NP constituted by elements among which there is no noun agreement instead of the NP composed by elements that harmonize themselves as to number; (ii) in the speach of this community, the target variable presents characteristics that show evidence of a change in progress in terms of Labov (l972). Given these assumptions, we conducted a quantitative analysis of a corpus of spoken language data from 27 (twenty seven) sociolinguistic interviews with a sample of speakers from three different social groups (medium, and low classes) divided into three age groups (young - 17 to 23 years; adult - 40-47 years; aged - above 60 years). These informants were selected considering their different levels of education: Primary Education (complete or not), High School (complete or not), and Higher Education (complete or not). Of the interviews with those 27 informants 1,297 NFs were selected; among them, 164 containing two non-nuclear elements (so that 1,461 tokens were subjected to quantitative analysis through the VARBRUL program). The results obtained showed that absence of concord in noun (ACN) occurs in 759 cases, or 52% of the analyzed data - which therefore means that the first hypothesis that guides this work is confirmed. Concerning the nine groups of factors initially taken, five of them were pointed out as having significant influence on the behavior of the target variable, and among the factors of these groups, the following factors were benefiting the ACN non-nuclear element of the NP: grammatical class, non-nuclear part of the NP: presence and absence of plural inflexion, gender, education, and social group of the informants. The quantitative results allow us to conclude that the target variable does not represent a change in progress at all, but it is characterized as a case of stable variable - this means that the second hypothesis above was rejected, confirming thus the conclusions of previous studies that states that in PB the linguistic variation related to concord is definitely internalized in the minds of speakers.