Setsuko Shirai | Aoyama Gakuin University (original) (raw)
Related Authors
The National Institutes for the Humanities (Inter-University Research Institute Corporation)
Uploads
Papers by Setsuko Shirai
Japanese often gives the signs of listening or agreeing by saying “soudesune.” Young people have... more Japanese often gives the signs of listening or agreeing by saying “soudesune.” Young people have used the shortened versions, “desune,” “dayone” and so on. In this paper, I attempted to assume the starting date for using these shorten forms in novels..
1 Since a word plus a particle (function word) has at most one pitch accent, the combination of a... more 1 Since a word plus a particle (function word) has at most one pitch accent, the combination of a noun plus a following particle is considered a prosodic word in Japanese (Poser 1984). 2 Japanese has other function words, too. However, it is hard to find the content words that consist of the same phonemes as the function words. 3 The second grammatical role of [de] indicates the end of a verb phrase.
Journal of The Acoustical Society of America, 2001
There is much debate about whether vowel centralization results from reduction of undershoot. The... more There is much debate about whether vowel centralization results from reduction of undershoot. The results of research conducted by Keating and Huffman (1984) indicated that Japanese vowels in prose were formed closer to the center of vowel chart than in words. However, they did not provide any statistical information and the information about the preceding consonants in the prose. Thus, there was a possibility that the preceding consonants led to the vowel centralization through undershoot. To address this question, I conducted research, in which Japanese function vowels [a, e, o] were compared with content vowels. The results showed that F1 of function /a/ following /g/ (average 609.8 Hz) was statistically lower than content /a/ (average 696.3 Hz) [F(1,65)=73.40, p<0.001]. However, there were no significant differences of F2 of /e/ following /d/ and F2 of /o/ following /t/ between function and content. At first, it appeared centralization played a role. However, close examination of the results indicated that vowel undershoot was the source of the /a/ centralization. Japanese function /a/ was statistically shorter than content /a/ and there was a significant correlation between duration and normalized vowel displacements for /a/ [Pearson's r=0.466, p<0.001]. This short duration caused the difference of F1 between content /a/ and function /a/.
When an English word is adapted as a Japanese loan word, some consonants are geminated. I conduct... more When an English word is adapted as a Japanese loan word, some consonants are geminated. I conducted research on gemination in a corpus of 3,399 Japanese loanwords derived from English. In my corpus, most geminates are developed from two source environments: 1) when a consonant is the last single consonant of a source word (i.e. word-final consonant), and 2) when a consonant follows a stressed syllable (i.e. ambisyllabic consonant). In both environments, gemination requires two conditions: the lax vowel condition and the singleton condition. The lax vowel condition requires that the vowel preceding the geminating consonant be lax. The singleton condition requires that the geminating consonant must not be part of a consonant cluster in the source words. Next I focused on the frequencies of gemination. My conclusion is that voiceless affricates and stops are geminated most of the time. In addition, the word-final consonants are more frequently geminated than the ambisyllabic consonants.
Japanese often gives the signs of listening or agreeing by saying “soudesune.” Young people have... more Japanese often gives the signs of listening or agreeing by saying “soudesune.” Young people have used the shortened versions, “desune,” “dayone” and so on. In this paper, I attempted to assume the starting date for using these shorten forms in novels..
1 Since a word plus a particle (function word) has at most one pitch accent, the combination of a... more 1 Since a word plus a particle (function word) has at most one pitch accent, the combination of a noun plus a following particle is considered a prosodic word in Japanese (Poser 1984). 2 Japanese has other function words, too. However, it is hard to find the content words that consist of the same phonemes as the function words. 3 The second grammatical role of [de] indicates the end of a verb phrase.
Journal of The Acoustical Society of America, 2001
There is much debate about whether vowel centralization results from reduction of undershoot. The... more There is much debate about whether vowel centralization results from reduction of undershoot. The results of research conducted by Keating and Huffman (1984) indicated that Japanese vowels in prose were formed closer to the center of vowel chart than in words. However, they did not provide any statistical information and the information about the preceding consonants in the prose. Thus, there was a possibility that the preceding consonants led to the vowel centralization through undershoot. To address this question, I conducted research, in which Japanese function vowels [a, e, o] were compared with content vowels. The results showed that F1 of function /a/ following /g/ (average 609.8 Hz) was statistically lower than content /a/ (average 696.3 Hz) [F(1,65)=73.40, p<0.001]. However, there were no significant differences of F2 of /e/ following /d/ and F2 of /o/ following /t/ between function and content. At first, it appeared centralization played a role. However, close examination of the results indicated that vowel undershoot was the source of the /a/ centralization. Japanese function /a/ was statistically shorter than content /a/ and there was a significant correlation between duration and normalized vowel displacements for /a/ [Pearson's r=0.466, p<0.001]. This short duration caused the difference of F1 between content /a/ and function /a/.
When an English word is adapted as a Japanese loan word, some consonants are geminated. I conduct... more When an English word is adapted as a Japanese loan word, some consonants are geminated. I conducted research on gemination in a corpus of 3,399 Japanese loanwords derived from English. In my corpus, most geminates are developed from two source environments: 1) when a consonant is the last single consonant of a source word (i.e. word-final consonant), and 2) when a consonant follows a stressed syllable (i.e. ambisyllabic consonant). In both environments, gemination requires two conditions: the lax vowel condition and the singleton condition. The lax vowel condition requires that the vowel preceding the geminating consonant be lax. The singleton condition requires that the geminating consonant must not be part of a consonant cluster in the source words. Next I focused on the frequencies of gemination. My conclusion is that voiceless affricates and stops are geminated most of the time. In addition, the word-final consonants are more frequently geminated than the ambisyllabic consonants.