Acquisition of intonational prominence in English by Seoul Korean and Mandarin Chinese speakers

This study investigated the production of intonational prominence contrasts in American English (AE) by Mandarin Chinese (M1 and M2) and Korean (K1 and K2) speakers differing in amounts of AE language experience. These languages differ in that English is a stress language where pitch accents are associated with stressed syllables in prominent words; Mandarin is a stress and tone language where prominence influences the extent of tonal changes in stressed syllables; and Korean is a nonstress language, where prominence is revealed through changes in pitch range and intonational phrasing.; Productions of word pairs differing in the location of stress (e.g., memorizes/memorial) were produced in statements and questions, and in sentential contexts where they were produced with sentence focus, before the sentence focus, and after the sentence focus. The fundamental frequency (F0) characteristics of target words and utterances are characterized, and are statistically analyzed in terms of F0 timing and F0 change. In addition, paradigmatic comparisons of duration in target syllables are statistically analyzed.; Results indicate that native language influences production of intonation patterns in English. F0 comparisons indicate that native English speakers produced different F0 patterns consistent with different pitch accents in target words in questions and statements, and produced pitch accents in only prominent target words. In contrast, nonnative English speakers consistently produced the same F0 patterns in target words regardless of the intonational context, and produced stressed syllables with higher F0 values than unstressed syllables in both prominent and less prominent words. These patterns were evident in the majority of productions by less and more experienced Korean- and Mandarin-English speakers. Duration comparisons indicate that native Korean speakers may have more difficulty acquiring reduction patterns than do Mandarin speakers. Both M1 and M2 speakers produced short unstressed syllables. For K1 and K2 speakers, this reduction strategy may be limited by the location of the unstressed syllable in a word. Producing longer stressed syllables posed no difficulty for any language group.; These results are discussed in terms of the Speech Learning Model, and conclusions are made regarding possible misguidings in ESL (English as a Second Language) instruction.