Capturing urban diversity through languages: Long‐term changes in multilingual residential neighbourhoods in the Helsinki Metropolitan Area

Abstract

Contemporary cities are becoming highly diverse on an unprecedented scale due to accelerating urbanization, international migration and mobility. Research on urban diversity commonly focuses on ethnicities and countries of origin, disregarding equally informative characteristics, such as information on languages and language use, which might reveal new aspects of urban diversity. In this paper, we analyse the development of urban diversity in the Helsinki Metropolitan Area between 1987 and 2019 using language information retrieved from the Finnish population register and examine the extent to which linguistic diversity captures urban diversity. We focus on two population groups, speakers of Estonian and Somali, to analyse how linguistic diversity and socioeconomic characteristics have changed in their residential neighbourhoods and to estimate to what extent these population groups have integrated into the host society. We measured linguistic diversity using Shannon entropy, performed spatial clustering analyses and used Markov chains to examine the likelihood of changes in linguistic diversity. We found clear spatial patterns and high spatial variability in the development of linguistic diversity, which shows that geographical space is a central component in linguistic diversity. Our results suggest that linguistic diversity is more likely to change in more segregated neighbourhoods and show that linguistic diversity is a suitable proxy for urban diversity. Understanding urban diversity is a key element for supporting social sustainability and cohesion in urban areas now and in the future. Mapping the spatio‐temporal and intersectional patterns of urban diversity is crucial for managing its negative effects, but more importantly, in leveraging its positive effects.

Sociolinguistic Variation in Urban Linguistic Landscapes

“…The register is updated yearly and features approximately 140 demographic variables covering every individual, including information on self-reported first language, family relations, educational and income levels, and approximate home location within a 250-metre grid. Combined with information on the first language, home locations allow the use of spatial methods to locate potentially interesting areas, such as grid cells with a high number of unique languages, while also tracking their changes over time (Väisänen et al 2023). However, the first language of an individual -assigned upon entry to the register -does not necessarily reflect the languages used in everyday life.…”