Perception of Japanese consonants by native speakers of American English

Abstract

Four native speakers of Japanese produced Japanese multi-syllable words and non-words, which include stops, fricatives, affricates, nasals and the liquid. Twelve native speakers of American English, who had never learned Japanese, heard their utterances and spelled out what they heard in English alphabet. What was of interest was whether native speakers of American English would perceive Japanese consonants in a way they are transcribed in English alphabet. The results revealed that Japanese stops were equated with generally the phonetically closest English stops, but voiceless stops were more likely to be equated with voiced stops of the same place of articulation than the other way around. Among voiced stops, /b/ and /d/ respectively were equated with /v/ and /l/. The word-initial /ts/, though words like "tsunami" have become part of English, was predominantly equated with /s/ rather than /ts/. The Japanese liquid, which is usually transcribed as /r/, was predominantly equated with /l/ rather than /r/. This agrees with results of previous research that demonstrate that English /r/ is more dissimilar from Japanese /r/ than English /l/ (See Aoyama et al. 2004).