Research Article: The geographical configuration of a language area influences linguistic diversity

Date Published: June 12, 2019

Publisher: Public Library of Science

Author(s): John L. A. Huisman, Asifa Majid, Roeland van Hout, Richard A. Blythe.


Like the transfer of genetic variation through gene flow, language changes constantly as a result of its use in human interaction. Contact between speakers is most likely to happen when they are close in space, time, and social setting. Here, we investigated the role of geographical configuration in this process by studying linguistic diversity in Japan, which comprises a large connected mainland (less isolation, more potential contact) and smaller island clusters of the Ryukyuan archipelago (more isolation, less potential contact). We quantified linguistic diversity using dialectometric methods, and performed regression analyses to assess the extent to which distance in space and time predict contemporary linguistic diversity. We found that language diversity in general increases as geographic distance increases and as time passes—as with biodiversity. Moreover, we found that (I) for mainland languages, linguistic diversity is most strongly related to geographic distance—a so-called isolation-by-distance pattern, and that (II) for island languages, linguistic diversity reflects the time since varieties separated and diverged—an isolation-by-colonisation pattern. Together, these results confirm previous findings that (linguistic) diversity is shaped by distance, but also goes beyond this by demonstrating the critical role of geographic configuration.

Partial Text

The diversity found across the world’s languages today is not the same as it was a hundred or 10,000 years ago, nor will it stay the same in the future. As the processes of diversification need time to run their course, we often find more diversity in areas where a language has been used for longer—compare, for example, English in the United Kingdom with English in Australia [1]. On top of this temporal dimension, we also see that linguistic diversity increases over geographical distance. Several patterns of linguistic diversity have been shown to exist, ranging from gradually accumulating differences [2], to more burst-like diversification [3]. The specific role that the geographical configuration of a language area plays in this process is less explored. The current study aims to investigate to what extent a cultural process such as language diversification follows the same patterns as a biological diversification. To do this, we investigate patterns of linguistic diversity in the context of an island setting by applying insights from population genetics.

It is clear that geography influences linguistic diversity, just as it influences biological diversity. However, the exact nature of this relationship in the context of languages is still poorly understood. Here we discovered that the geographical configuration of a language area affects the role of two known diversification processes: dispersal and colonisation history. After a cluster analysis based on linguistic distance measures confirmed the legitimacy of Ryukyuan—spoken across isolated island clusters—as a language group distinct from Japanese—spoken across a connected land—we examined the relationship between geographical distance and linguistic diversity, as well as time since divergence and linguistic diversity in these two language areas. As expected, linguistic diversity in both language areas increased with larger geographic distances, and with increased time since speech communities separated for the Ryukyuan area. Importantly, we found that the effect of geographic distance was stronger for Japanese, while the effect of time since divergence was stronger for Ryukyuan—a result of two different processes that have shaped linguistic diversity.

To conclude, we have shown that cultural processes—language diversification—are influenced by geography in ways similar to biological processes—species diversification. We examined the role of geographic configuration in diversification and showed that: (I) mainland languages display a typical isolation-by-distance pattern, with gradually increasing diversity over geographic distance, as a result of the higher potential for sustained contact, while (II) island languages display a typical isolation-by-colonisation pattern, where diversity is a reflection of time since divergence, as a result of limited contact due to the geographic isolation of islands. Language variation and change is, of course, influenced by other (historical and socio-political) factors as well, and a more global and multi-dimensional concept of distance—comprising spatial, temporal, and social factors—is needed to help us understand patterns of language diversification. Our results show that the geographical configuration of a language area is one important component of a more comprehensive distance concept to explain language variation and change.