The Statistical Inference of Sentence Patterns and Its Implication for Oral Interpreters

• Main Authors: Che-chang Kang, 康哲彰
• Other Authors: Ping-yen Lai
• Format: Others
• Language: en_US
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/95983412069405694440

碩士 === 國立彰化師範大學 === 翻譯研究所 === 95 === In this study, the researcher uses the quantitative and qualitative approaches to investigate the sentence patterns of speech texts and written texts. The emphasis of the investigation is placed on the AWS (Average Words per Sentence) and Standard Deviation of sentence length in speech texts and written texts. For the quantitative approach, the researcher uses the inferential t-test to investigate: (1) the difference in the AWS between speech texts (30 pieces) and written texts (30 pieces) when the speakers and writers are different; (2) the difference in the AWS between speech texts (25 pieces) and written texts (25 pieces) when the speaker and the writer are the same individual; (3) the difference in the AWS between scholars’ and politicians’ speech texts (respectively 35 and 38 pieces); (4) the difference in the Standard Deviation of sentence length between scholars’ and politicians’ speech texts (respectively 35 and 38 pieces). Then, the researcher gives a qualitative discussion to find out the possible factors behind the statistical results. Finally the researcher draws some implications of the statistical inference of sentence patterns for oral interpreters from three perspectives: strategy; preservation of style; psychological anticipation. The major findings of this study are: (1) there exists a statistically significant difference in the AWS between speech texts and written texts when the speakers and writers are different (t=6.875, p<0.001); (2) there exists a statistically significant difference in the AWS between speech texts and written texts when the speaker and the writer are the same individual (t=4.769, p<0.001); (3) there exists a statistically significant difference in the AWS between scholars’ and politicians’ speech texts (t=5.183, p<0.001); (4) there exists no statistically significant difference in the Standard Deviation of sentence length between scholars’ and politicians’ speech texts (t=0.983, p>0.05). Overall, it is concluded that the sentence patterns in speeches and writings are critical in reflecting the speakers’ and the writers’ distinctive styles. Therefore, in doing interpreting the oral interpreter should make sure proper strategies are applied to preserve these distinctive styles. In addition, the oral interpreter could engage into the psychological anticipation of the speakers’ backgrounds and intentions based on the different sentence patterns used by the speakers. This line of discussion is put forward under rather theoretical manner in this study, but has served to point out the indispensable anticipation ability that an oral interpreter should be equipped with in doing interpreting.