Análise do desenvolvimento do ritmo de inglês-L2 de aprendizes brasileiros

Abstract

The aim of this study is to describe and discuss the development of L2 English rhythm by Brazilian learners through rhythmic metrics and prosodic-acoustic parameters that characterize the oral production of these learners at different stages of L2 development. This study adopts a dynamic perspective of L2 development (DE BOT; LOWIE; VESPOOR, 2007; DE BOT, 2008; LARSEN-FREEMAN, 1997; LARSEN-FREEMAN; CAMERON, 2008; LOWIE; VERSPOOR, 2015) and a multidimensional and gradient approach to speech rhythm (CUMMING, 2010; FUCHS, 2016; LOUKINA et. al., 2011; SILVA JR.; BARBOSA, 2019b). Four corpora of oral production were analyzed. Five Brazilian learners of English-L2 were recorded reading a text in English at the beginning of their college studies in English Language Teaching, and again four semesters later, after having taken two English phonology courses. They were also recorded reading a version of the text translated into Portuguese. Besides the learners, five native speakers of North American English were recorded reading the same text in English. Thus, this study also seeks to evaluate possible effects of explicit pronunciation instruction on the development of the rhythm of English by the learners. Data were manually segmented into vowel units (V), consonant (C), vowel-vowel (VV), sentences (S) and higher prosodic units - chunks (CH) for the extraction of rhythmic metrics in PRAAT (BOERSMA; WEENINK, 2021), and the script Metrics&AcousticsExtractor (SILVA JR.; BARBOSA, 2019) was used to extract 30 rhythmic metrics and 12 acoustic parameters. Data were then statistically treated via R (R CORE TEAM, 2021) through the implementation of mixed-effects regression models, adopting language and semester as predictor variables, and rhythmic metrics and acoustic parameters as response variables. Results placed Brazilian Portuguese and English-L1 in different rhythmic spaces, as predicted by the literature; in the durational dimension, the metrics positioned the English-L2 of the first recording far from both English-L1 and Brazilian Portuguese; in the f0 dimension, however, the acoustic parameters placed the English-L2 of the first recording closer to Brazilian Portuguese. In both dimensions, the English-L2 of the subsequent recording was closer to English-L1, suggesting a developmental route towards the target language. The results also suggest positive effects of the explicit teaching of pronunciation.