A aquisição das consoantes laterais do português europeu por aprendentes chineses

Abstract

The present study examined the production of European Portuguese (EP) lateral consonants by 14 Chinese learners, through a picture naming task eliciting the target segments in all possible syllable and word-level positions. Our results illustrate that /l/ is stable in singletons (100% target-like) due to the positive transfer from Mandarin Chinese. However, it is very often vocalized in codas (only 16.7% target-like production, [ɫ]), which might be attributed to a phonetically based tendency (Graham, 2017; Johnson & Britain, 2007). The high accuracy (97% target-like) of /l/ in onset clusters, an absent structure in the L1, can be the result of the heterosyllabic nature of EP obstruent-liquid sequences (Veloso, 2006) or of the association of two segments to a single skeletal position, which was also argued as an intermediate stage in EP L1 acquisition (Freitas, 2003). /ʎ/ is still in acquisition (52.4% target-like), and is often produced as an L1 category [lj], due to acoustic and articulatory similarity.

Cross-Linguistic Interaction Between Phonological Categorization and Orthography Predicts Prosodic Effects in the Acquisition of Portuguese Liquids by L1-Mandarin Learners

“…To return to our first research question, we hypothesized that the previously-reported L2 prosodic effect on the acquisition of EP /l/ and /ɾ/ by L1-Mandarin learners [1,2] could be accounted for by CLI. The results of our delayed imitation task with naïve Mandarin listeners largely replicated the modifications employed by L1-Mandarin learners, as shown in Table 1.…”

“…Recent research [1,2] has shown that second-language (L2) acquisition of the European Portuguese (EP) lateral /l/ and rhotic /ɾ/ by L1-Mandarin learners is conditioned by prosodic position: learners almost never mispronounce EP /l/ in onset, but in coda frequently vocalize it as [w]. EP /ɾ/ is replaced by the learners with [l] in onset, while in coda they delete the segment, insert a schwa (and thus create an onset), or substitute it either with [l], a coronal stop or the Mandarin rhotic [ɻ].…”

L2 speech learning of European Portuguese /l/ and /ɾ/ by L1-Mandarin learners: Experimental evidence and theoretical modelling

“…Lin, 2001;Duanmu, 2005;Lin, 2007), since there is no phonological motivation for an underlying fricative /ʐ/, which would be the only voiced obstruent in the Mandarin phonological inventory and would introduce a novel phonological contrast [± voice] to Mandarin (Duanmu, 2007). 8 The phonetic distinction discussed here is restricted to intervocalic position, because a previous study (Zhou, 2017) demonstrated that L1-Mandarin learners do not confuse the EP lateral and the tap in coda position.…”

“…Despite many similar observations (e.g. Martins, 2008;Espadinha & Silva, 2009), systematic research on this notorious L2 difficulty is a relatively recent enterprise (Zhou, 2017;Liu, 2018;Cao, 2019;Vale, 2020). These experimental studies have revealed that the complexity of this L2 speech learning process goes beyond what was documented in previous descriptive studies, namely, the confusability between /l/ and /ɾ/ (Batalha, 1995;Martins, 2008), since L1-Mandarin learners' acquisition of this novel phonological contrast seems to be shaped by the interaction between different prosodic positions, speech modalities and representational levels.…”

“…The documentation of Chinese learners' struggle with EP /l/ and /ɾ/ can be traced back to Batalha (1995). It comes as a surprise, though, that it was not until recently that this notorious difficulty was investigated systematically in experimental studies (Zhou, 2017, Liu 2018Cao, 2019;Vale, 2020).…”