Perception of Sound Symbolism in Mimetic Stimuli: The Voicing Contrast in Japanese and English

Loading...
Thumbnail Image

Date

2018-10

Journal Title

Journal ISSN

Volume Title

Publisher

Ohio State University. Libraries

Research Projects

Organizational Units

Journal Issue

Abstract

The current study investigated sound symbolism in Japanese mimetic stimuli. We examined whether the voicing contrast in consonants (/d, g, z/ vs. /t, k, s/) affects perception both in Japanese native speakers and in English native speakers. Stimuli were evaluated on 4 different dimensions: size (big-small), shape (round-spiky), and two evaluative dimensions (good-bad, graceful-clumsy). The voicing on the stimuli was manipulated, creating a continuum from voiced to voiceless endpoints in order to examine the categorical nature of the perception. The current study found that both Japanese and English speakers tended to associate voiced sounds with largeness, badness, and clumsiness and voiceless sounds with smallness, goodness, and gracefulness. However, for the shape dimension, English speakers only tended to associate voiced stop consonants with roundness and voiceless stops with spikiness. The present results show systematic sound symbolic relationships in terms of voicing for Japanese and English.

Description

Keywords

Sound Symbolism, Japanese, Mimetics, Consonantal voicing

Citation

Nakata, Kotoko. "Perception of Sound Symbolism in Mimetic Stimuli: The Voicing Contrast in Japanese and English." Buckeye East Asian Linguistics, vol. 3 (October 2018), p. 53-63.