The Acquisition Of French Stops By Chinese Learners And The Dialectal Backgrounds' Influence On Their Acquisition

This study examines the production and perception of French stops by Chinese learners of French,using Voice Onset Time(VOT)as an acoustic cue and dialectal background(i.e.,with voiced stops or not)as an essential factor of variation.Research methods like questionnaire,acoustical experiments,perceptional experiments and statistics are adopted.By collecting French stops' production and perception data of thirty Chinese learners of French,this study examines the acquisition of participants as a whole and as two groups with different dialect backgrounds.The production data is collected from stops in citation forms and carrier sentences,while perception data is gathered from two identification tasks with their stimuli VOT ranging from-150ms~150ms and-50ms~50ms.The analysis shows that participants as a whole tend to assimilate French voiced and voiceless stops to Mandarin unaspirated ones.For French voiced stops,participants make two kinds of mistakes,namely,devoicing and exaggerated speech imitation.Meanwhile,analysis shows that Chinese learners' perception boundary for French stops is heavily influenced by their mandarin experience,a result that corresponds to the “single category” in the Perceptual Assimilation Model framework.In the identification task with narrower stimuli VOT range,Chinese learners fail to show a clear perception boundary for bilabial and alveolar stops,while the perception boundary for velar stops are similar to that of French natives',which may be accounted for by the perceptual magnet effect.In terms of the influence of different dialect backgrounds on French stops' acquisition,production comparison of the two groups of learners shows that learners from a dialectal background with voiced stops have a higher voicing rate for voiced stops,but have no advantage in its accuracy.For perception data,learners from a dialectal background with voiced stops are relatively closer to native speakers of French in the first identification task,though significant differences also exist.However,in the identification task with narrower stimuli VOT range,the two groups show no significant differences.The result indicates that Chinese learners of French stops tend to confuse voiced stops with voiceless ones,which can be accounted for by “equivalence classification”in SLM and “single category” in the PAM framework.