An Emerging Gender Difference in Japanese Vowel Devoicing 185is complex, not direct and ¿ xed, and that there are many factors to be considered
when we talk about women’s language: sexual orientation, social and contex-
tual diversity, and the speaker’s beliefs about language use. So, young females’
increased use of men’s language may not mean women’s language is disappear-
ing. However, it is still true that there is a new meaning attached to vowel devoic-
ing among younger speakers as a group, as the results of this study show.
If we take the hypothesis suggested here seriously, our next question is, “How
much devoicing is standard and how much is nonstandard?” If vowel devoicing
is a standard feature and young males are devoicing more and young females
devoicing less, we might think the standard amount of devoicing is somewhere
in the middle, approximately where the rest of the groups are. If standard speak-
ers feel (subconsciously, of course) that Kansai (western Japan) speakers are
not devoicing enough, are the female speakers taking a risk of sounding like
Kansai speakers? Probably not. Young females are attracted to standard speech,
and those young females sound very standard to the ears of standard speakers.
They are within the territory of standard devoicing. One might argue that there
are other features that distinguish Tokyo speakers from Kansai speakers, such
as pitch accent, although vowel devoicing is still a good parameter (cf. Morris
2003). The questions here is, then, how much devoicing do Kansai speakers
have compared to Tokyo speakers? Are the environments in which devoicing
occurs different in the two major dialects of Japanese? The answer to these ques-
tions has to wait until another production study on Kansai speakers is done.
In conclusion, a detailed acoustic study of Japanese vowel devoicing
revealed the emerging gender difference among younger speakers of the
Tokyo dialect. If we had not considered this factor, Japanese women’s lan-
guage may have appeared to be on the way to merging with men’s language,
but that is not the case at the phonetic level, at least in terms of the use of
vowel devoicing. A study of Kansai speakers’ vowel devoicing rates is needed
to determine the standard amount of vowel devoicing in Japanese. Moreover,
more systematic study is needed of younger speakers’ speech on multiple
levels (lexicon, phonetics, morphology, syntax, etc.) to reveal the social cor-
relates of current language change in Tokyo Japanese.
References
Beckman, Mary and Atsuko Shoji. 1984. Spectral and perceptual evidence for CV
coarticulation in devoiced /si/ and /syu/ in Japanese. Phonetica 41: 61–71.
Haig, John H. 1990. A phonological difference in male-female speech among teenagers
in Nagoya. In Ide and McGloin (eds), 5–22.