A Reader in Sociophonetics

Regional Stereotypes and the Perception of Japanese Vowel Devoicing 199

5.3 Gender of the speaker

Tables 7.11, 7.12, and 7.13 show the results of perception of voicing variants
by gender of the speaker. The results with Tokyo devoiced-nondevoiced com-
bined data and Kinki devoiced-nondevoiced separated data are not signi¿ -
cant. Tokyo results for devoiced and nondevoiced show that women’s voices
promote the expected responses; that is, female devoicers tended to be judged
as Tokyo and female nondevoicers tended to be judged as non-Tokyo. This
suggests that Tokyo respondents are more sensitive to the stereotype that
women are more likely users of the standard. In contrast, the Kinki results
show that men’s voices promote expected responses. Respondents may asso-
ciate a covert prestige idea with men’s voices, perhaps assuming that male
nondevoicers are tough and/or cool-sounding Kinki persons and that male
devoicers are not.

Gender Weight

Female 0.542

Male 0.444

Gender Weight

Female 0.588

Male 0.436

Gender Weight

Male 0.521

Female 0.479

Voic i ng % / We ig ht

Nondevoiced 54.69/0.521

Devoiced 52.17/0.483

Table 7.10 Kinki Results by Voicing Variants

Table 7.11 Results by Gender of the Speaker (Tokyo, Devoiced)

Table 7.12 Results by Gender of the Speaker (Tokyo, Nondevoiced)

Table 7.13 Results by Gender of the Speaker (Kinki, Combined)