Early stages of adult L2 acquisition of Mandarin :an organic grammar approach

Abstract

There has been debate over whether the theory of Organic Grammar (OG) (Vainikka and Young-Scholten 1994, 1996a and 2011) can predict L2 development in languages other than European ones. Specifically, OG argues that second-language learners move from an initial bare VP stage to intermediate IP-level stages to a final CP stage, regardless of their L1 or L2 status. Moreover, the order of acquisition of the relevant elements is the result of the interaction between Universal Grammar and the target language input. Counter-arguments have been found in Schwartz and Sprouse (1994, 1996), Epstein, Flynn and Martohardjono (1996), Haznedar (1997, 2003), Lardiere (1998, 2008) etc. with discussion focused on L1 functional transfer and the stage-like development of functional projections. This thesis investigates the extent to which the theory of OG can explain aspects of Mandarin Chinese acquisition by native English-speakers. Extending existing proposals, I hypothesise a clausal structure model for Mandarin, which includes the functional projections AspPl, BaP, BeiP, AspPm, NegP, AspPh, headed respectively by lel/guo/zhe, ba, bei, zai, bu/mei and leh respectively. Based on OG, L2 Mandarin learners are predicted to start with a bare VP and demonstrate word order consistent with their first language, and to develop over time in the following functional sequence: AspPl > BaP [lower IP stage] > BeiP > AspPm> NegpPh > AspPh [upper IP stage] based on access to UG and the input. Oral production and grammaticality judgement data were collected through administering repeated multi-task tests to eight instructed English-speaking learners of Mandarin during their first and their second years of university study between October 2014 and June 2015. Emergence and accuracy criteria were used to analyse the production data, while the grammaticality judgement data were analysed through the repeated measures of the SPSS General Linear Model. The thesis presents three main findings. Firstly, when the word order in the verb phrase is different in English and Mandarin, the learners in this study use the order similar to their L1 English, thereby indicating the learners’ transfer of their L1 word order. Secondly, there was largely a distinctive stage-like development based on hypothesised hierarchical functional projection structure; nonetheless, variability was also found in the development of individual learners. Thirdly, the acquisition results could be attributed to a great extent to the interaction between Universal Grammar and target language input; although the role of cognitive factors like memorisation should not be ignored. Overall, the results revealed the predicted stages; thereby supporting OG’s explanatory value by extending it to the analysis of L2 Mandarin.