The Production Of Tone 3 By American Mandarin Learners

Variations in Mandarin Tone 3(T3)have always been a hot topic in linguistic studies.In the study of Mandarin tones by second language(L2)Mandarin learners,few studies have focused on the influence of phonological environments,the prosodic hierarchy or learning time on the acquisition of Mandarin T3.This study is to explore such influences by taking into consideration the study on Mandarin tones in Duanmu(2007)and Huang & Liao(2011)and the study on metrical words and prosodic hierarchy in Feng(1996)and Kuang & Wang(2006).Recordings of 24 native American English speakers(gender balanced)are selected from the Global TIMIT(L1＿English)corpus to study how phonological environments within and between metrical feet influence the production of T3,and how learning time influences the production of T3.The K-means clustering and ANOVA test are used to analyze the normalized pitch.Findings of this study are as follows.(1)Within the metrical foot,phonological environments influence the T3 contour produced by American learners.In the combinations of “T3+T2” and “T4+T3”,L2 learners tend to produce falling or falling-rising contour,while in other bisyllabic tone combinations,L2 learners produce one more pattern,i.e.,the rising contour.(2)Between metrical feet,the T3 in the disyllabic words can be influenced by the tone preceding and following the T3 in the adjacent feet.In “TX+(T3+T3)+TX” combination,the first T3 will be significantly influenced by the preceding tone,and the second T3 will be significantly influenced by the following tone.In “TX+(T3+T2)”,T3 is greatly influenced by TX,the preceding tone.(3)The learning time of Mandarin has no significant influence on the pitch range.That is,there is a great fluctuation in the T3 pitch range for an L2 learner given controlled phonological environments and prosodic hierarchies.The learning time has significant influence on the T3 contour in “T1/2/4+T3” and“T3+T4” combination.That is,L2 learners whose learning time is more than three years tend to produce T3 in falling-rising contour,while those with less than three years tend to pronounce T3 in falling or rising tone.In conclusion,it’s found that the phonological environments within foot or between feet,and the learning time,have different influence on the production of T3 and T3 variations.This study also finds that contours of T3 and T3 variations produced by American learners are of large difference compared with the rules of Mandarin tone variations proposed in Duanmu(2007)and in Huang & Liao(2011);it’s hard for American learners of Chinese to pronounce T3 in the way of native Chinese speakers.After the defining of metrical feet according to the methods proposed by Feng(1996),it’s found that prosodic hierarchies will influence T3 production,which is in accord with the findings in Kuang & Wang(2006)on native Chinese speakers.