Ethnic segregation on linguistic landscapes

Abstract

This work presents a novel approach to studying ethnic segregation from the perspective of linguistic landscapes. Numerous street-level images accumulated over the last two decades have enabled the exploration of linguistic landscapes at a larger scale than ever before. Since the prevalence of a specific language in a public space implies the linguistic group inhabiting the area, its careful evaluation can reveal the degree of segregation between linguistically different ethnic groups. To demonstrate the effectiveness of the proposed approach, we applied it to the linguistic landscape of Seoul, South Korea. Using a large set of street-level images collected from an online map platform, we measured the levels of segregation between Korean and Chinese signs from 2010 to 2018. The levels of segregation on the street-level images were different to a certain extent from those of residential segregation. While residential segregation gradually increased between 2010 and 2018, except for 2011, more fluctuations were observed in linguistic segregation. This finding is likely because a linguistic landscape is shaped mainly by advertising signs, banners, and billboards in commercial areas, and such commodified urban spaces change more dynamically to attract inhabitants and visitors. These results suggest that the proposed approach can offer an alternative way of understanding the complex socio-demographic phenomenon from a new perspective, as with other emerging data sources in the era of big data.