Anti-transfer in L3A of Portuguese

Abstract

A growing body of research in third language acquisition has demonstrated the possibility of backward cross-linguistic transfer, specifically L3L2 transfer (Hui, 2010; Cheung, Matthews & Tsang, 2011; Tsang, 2014). Among these studies, Matthews et al. (2014) suggest that L3L2 transfer could actually counterbalance previous transfer from L1 to L2, which they referred to as “anti-transfer effect”. In their study, L3 German learners (L1-Chinese and L2-English), compared to L2 English learners (L1-Chinese), used significantly fewer uninflected verb forms to describe past events in their L2 English, a type of error attributed to negative transfer from their L1 Chinese which lacks tense markers. This reduction of uninflected verb forms may be due to the systematic tense inflection in German enhancing L3 learners’ morphological awareness for tense marking. In light of this possibility, this project sets out to investigate the influence of L3 on previous L1L2 transfer in the domain of English third person singular agreement by looking at L3 learners of Portuguese (L1-Chinese and L2-English) in Macao, a previous Portuguese colony. Adopting longitudinal and cross-sectional methods, it is hypothesized that after one year of Portuguese learning, learners’ use of uninflected verb for third person singular in English will be reduced and this effect will increase in proportion to learners’ L3 proficiency. In a pilot study comparing a L3 group with 2 years of Portuguese learning experience and a L2 group with no learning experience in Portuguese, it is found that numerically, the L3 group showed an advantage of the use in English third person singular over the L2 group, in both the writing task and the grammaticality judgment task. The result from the pilot suggests L3 influence in accordance with our hypothesis. A DST approach (Jessner, 2008) will be discussed, focusing on the role of metalinguistic awareness in the development of multilingualism.