Intonation Patterns in English Tag Questions of Japanese Speakers of English as a Second Language

Ten female native speakers of American English (NS) and ten Japanese females who speak English as a second language (NNS) were recorded reading a text containing emphatically-assertive tag questions. Evidence was sought to determine whether or not native speakers can correctly identify these utterances as produced by native or non-native speakers on the basis of intonation alone. The tag questions were low-pass filtered to remove segmental content, then played to ten native speaker jurors who judged them as native or non-native. Judges correctly identified NSs at 73.5% and NNSs at 71.3%. This study further investigates what differences exist between the two groups. Utterances that were unanimously correctly identified were analyzed using the ToBI transcription system. Areas of distinct differences between the two groups were subjected to acoustic examination of frequency, duration and intensity. NS utterances were more homogeneous than were NNS utterances.