In Ibuki Island, Kagawa Prefecture, a "kako-shiki (declining)" accent is observed by some scholars. Judged with the aural impressions, the previous studies have stated that this accent pattern has subtle declining tone from the second mora to the third mora. In this article, we measure frequency movements of this accent and compare them with those of the words in the second mora nucleus type accents. The result is as follows. The difference clearly exists between the kako-shiki accent and the other accents on Ibuki Island. And the peak position of frequency movement does not always exists between the second mora and the third mora. In addition, we collect samples of the small declining tones in three different places in the northeastern Shikoku area. We try to extract frequency movement patterns in each mora, and to characterize these patterns.
This paper describes the accentual system of the Sanagishima Dialect, centering on (1) the opposition and neutralization of two registers (low-level register and high-level register), (2) phonological interpretations of tonal patterns which occur in quasi-complementary distribution conditioned by vowel aperture and morpheme boundary, and (3) accent changes in progress which influence the interpretations of the accentual system.
This manuscript concentrates heavily on the fact that the accent in the Tanohama dialect of Seiyo City in Ehime Prefecture is determined by the moraic structure of the word. It also touches briefly upon phonetic factors such as the notion that sounds in the "ga" row of the Japanese syllabary are sometimes pronounced with a [ŋ] and that sounds in the "ka" and "ta" rows of the syllabary are sometimes produced with a sound that is close to the voiced [g] [d].
This paper describes vowel coalescence in the dialects of Shimanami-Kaido (a highway between Hiroshima and Ehime prefecture) region. The discussion includes not only a description of the phenomenon but also a difference of region and age. The result is that vowel coalescence is easy to occur in the environment which include morpheme boundary, for example /V+i/ or /V+u/(V=vowel). From the viewpoint of region, it is easy to occur relatively in the side of Hiroshima. And there is a difference between the older, middle age and the younger age. The younger age often uses the common forms.
This paper describes on-glide nasal [d] and [g] at initial and medial positions in Shikoku dialects. Acoustic analysis was used to examine on-glide nasal [g] at initial position in Kochi, Tokushima and Kagawa dialects and it was found that this phone occurs in Kochi dialects but not in Tokushima and Kagawa dialects. Language geographical analysis showed that on-glide nasal [d] and [g] at medial position occur in Tokushima dialect, but have almost disappeared except in isolated areas. Analysis also showed that, in Tokushima dialect, on-glide nasal [g] at medial position tends to disappear faster than on-glide nasal [d].
[ŋ] before vowels exist in various dialects of Japan. It shows wide range of distribution all over Japan, including Ryukyuan islands. In Tohoku area, voiced plosives and affricates before vowels are pre-nazalized. This paper treats velar nasal consonant as a variation of pre-nasalized voiced velar plosive, same as those pre-nasalized voiced bilabial and dental consonants. Some of the results of researches show the narrower distribution of pre-nasalization, which means that the degree of nasalization is decreasing according to the age of speakers of one dialects. The result of research in Kochi dialect on pre-nasalization shows the degree how the pre-nasalization decrease as the age of informants become younger.
A previous study (Mugitani et al. 2004) indicated that 8-month-old Japanese infants match the phonetic information of vowel /a/ to its respective visible lip articulation, while such audiovisual matching of lips and voice was not evidenced for vowel /i/. This study aimed at further verifying the development of audiovisual lip-voice matching of vowel /i/ in Japanese infants. Experiment 1 indicated that at 11 months Japanese infants could match vowel /i/. Experiment 2 suggested that Japanese 8-month olds still show no evidence of lip-voice matching of vowel /i/, although the familiarity of articulation manner was controlled. Longitudinal analysis on a child-mother speech database revealed that Japanese children at any age experience vowel /i/ less frequently in both perception and production. These findings indicate that, in Japanese infants, lip-voice matching of vowel /i/ develops between 8 and 11 months of age; the development is possibly effected by language-specific visual and motor experience of vowel articulation.