|
Intonation is the final form of multiple communicative functions encoded in FO and duration via target approximation. The main goal of this thesis is to investigate and compare the prosodic encoding of focus, and how focus and question are co-encoded together in three languages in China systematically and theoretically, which were experimentally studied using question-and-answer paradigm, the languages included Wa and Deang (Mon-Khmer), Tibetan (Tibetan-Burma). Moreover, the perception experiment of Tibetan is reviewed, too. Results from acoustic analyses showed that:[1] In Deang and Wa, speakers do not expand on-focus pitch range and not increase FO in any of the focus condition, duration and intensity changed but do not exhibit any statistical significance as a function of focus. There is no post-focus pitch compression(PFC) in any focus condition.[2] In Tibetan, all speakers raised pitch, duration and intensity of focused words, and lowered pitch,duration and intensity of post-focus words. There is about 90% focus recognition rate in medial focus condition, but in initial and neutral focus conditions, focus recognition rate is only about 40%.[3] The sites of focus don’t affect the intonation contour of question both in Deang and Wa. Across all four focus conditions, the differences in pitch between statement and question in Deang mainly lie in the last syllable, the question contour in last syllable tends to rise sharply. However, as to the Wa question contour, the whole pitch level is raised from the beginning of the sentence.In the end of the paper, on the basis of the conclusion of Deang,Wa and Tibetan, the paper also touched on the prosodic encoding of focus in other languages such as Uygur, Nanchang dialect (Gan), Yi,Taiwanese,Taiwan Mandarin,and Cantonese and so on. At the same time, the relationship between PFC and some other factors such as language family and the origin of language is discussed.Moreover,the issue of how focus and question are co-encoded together is also discussed, then the results of Deang, Wa, Beijing Dialect and Uighur are compared. The conclusions from the comparision support the "Paralle Encoding and Target Approximation" (PENTA).
|