182 Terumi Imai
- Results
4.1 Speech style
The results of the statistical analyses show that speech style is a signi¿ cant
factor in Japanese vowel devoicing. Respondents devoiced most in conversa-
tion, and more in reading passage than in word list. In other words, the more
casual the speech style, the more likely devoicing occurs. This con¿ rms the
results obtained in Yuen 1997. Table 6.1 and Table 6.2 show the results of a
Goldvarb run for speech style for the vowels /i/ and /u/ respectively.
Table 6.1 Speech Style for /i/—Goldvarb
Speech Style Weight Number (devoiced/total)
1Conversation0.816 1585/1785
2Reading passage0.486 097/1732 1
3Word list0.3351799/3740Table 6.2 Speech Style for /u/—Goldvarb
Speech Style Weight Number (devoiced/total)
1Conversation0.742 1585/1785
2Reading passage0.590 097/1732 1
3Word list0.3301799/3740In Table 6.1, only conversational style promotes the devoicing of /i/, and
the word list and reading passage styles demote it, even though devoicing is
more likely to occur in reading passages than in word lists. In Table 6.2, both
the conversational and reading passage styles are promoters for the devoicing
of /u/, and only the word list style demotes it. However, the general order is the
same: the more casual the style, the more likely devoicing occurs.
4.2 Age and sex
One surprising result of the current study is the interaction between age and
sex. When the two categories were cross-tabulated, there was a signi¿ cant
difference between young males and young females, but there was little sex
difference between other age groups. Younger males devoice most, younger
females least, and the rest fall in between, as shown in Figure 6.3 for /i/ and
Figure 6.4 for /u/.