Production of the Taiwanese Southern Min tones by L1 Mandarin speakers

Abstract

This study compares the Taiwanese Southern Min (TSM) tones produced by L1 Mandarin speakers and L1 TSM speakers in Taiwan. TSM has seven lexical tones with complex tone sandhi rules while Mandarin only has four lexical tones with a Tone 3 Sandhi. The acoustic study shows that the L1 Mandarin group produced all seven tones with pitch registers and contours similar to the L1 TSM group, including the uncategorized mid-level tone (Tone 7) and two short tones (Tone 4 & 8). In addition, the L1 Mandarin speakers produce accurate sandhi tones at non-XP-final positions. They had no difficulty applying the tone sandhi rules, suggesting that they might have stored the sandhi forms lexically. The further comparison to L1 Mandarin group’s Mandarin tones suggests that the rising and the falling tones were assimilated to their native ones. However, they struggled to reduce the duration of the two TSM short tones to the same degree as the L1 TSM speakers, which suggests that durational differences were more difficult than the pitch for them to acquire.