Incomplete lexical pitch accent representation in the standardization of accentless dialects in Japanese: an approach through speech perception experiments.

Most Japanese dialects, including the standard Tokyo variety, employ a lexical tonal system, whereas certain dialects, referred to as accentless dialects, lack such a system. Although dialectological research has frequently suggested that younger speakers in accentless regions have acquired the standard tonal system through dialect standardization, evidence from perception studies indicates that these distinctions may not yet be fully established. The present study investigates whether the acquisition of the standard tonal system among speakers of accentless dialects is complete. We employed both evaluations of accentless dialect speakers' speech production by naive listeners and a sequence recall task designed to assess phonological contrasts in speech perception. Our findings demonstrate that the speech production of accentless dialect speakers closely resembled that of standard Japanese speakers and was often indistinguishable to naïve listeners, consistent with previous claims of complete standardization. However, the perception experiment revealed significantly poorer performance among accentless dialect speakers in discriminating lexical pitch contrasts. These results suggest that the standardization of accentless dialects remains incomplete at the perceptual level, despite the apparent convergence in production. Moreover, the sequence recall task proved to be an effective tool for identifying subtle cases of incomplete acquisition in the standardization of dialects.