Phase 2: Theme A - Demographic processes and cultural change: Project A010|
Demographic models of language shift
Anne Kandler (AHRC CECD, Institute of Archaeology, University College London)
PROJECT FUNDING: 50% Leverhulme Trust - Early Career Fellowship;
50% AHRC CECD
Cards Against Humanity
Models of the population dynamics of language shift are essential tools for contemporary language planning. It is estimated that half of the world's living languages will be lost during the next century. This project uses mathematical models of language competition to predict the extinction or persistence of an endangered language in a bilingual community, and will help language planners judge the efficiency of proposed interventions. The initial focus will be on cases with very good existing census data (Gaelic in Scotland; French in Canada; Welsh in Wales). The model will then be applied to less fully-documented contemporary and historical situations.
FINAL PROJECT REPORT:
The main focus of this project was to analyse the relationship between demography and language shift. We developed a model which tracks temporal changes in numbers of speakers who are monolingual in the competing languages, and in the size of the bilingual subpopulation. The extent of these changes is determined by demographic factors, such as population growth rate and spatial dispersal/interactions, and by linguistic factors which define the social and economic advantages of shifting languages. The main findings are that these demographic processes and the initial distributions of speakers of the different languages can all influence the shift dynamic significantly. In a spatially and temporally constant environment the extinction of one language (which may not be the high-status language) is inevitable. However, coexistence between languages of different status can be achieved given some measure of spatial segregation. Further, we examine the role of diglossia, the different use of language in different social niches, on the shift dynamic. We found that such a social segregation mechanism can lead easily to a stable coexistence of languages between the subpopulation speaking the high-status language and the bilingual subpopulation.
In a first case study we explored the dynamic of Welsh-English and Gaelic-English competition in Wales and Highland Scotland respectively. Here we could use national decennial census data, whose availability extends back to the late nineteenth century and which contains information about the number of monolingual and bilingual speakers in different regions. We estimated the model parameters partly from external sources and partly by calibrating them to the data and obtained a convincing fit between our model and data. However, the most exciting result of this case study was that we were able to calculate the number of monolingual English speakers who now need to become bilingual every year to stabilize the bilingual English- and Gaelic-speaking community in the Highlands at its current level. This means that by using census data and projecting ahead from a time series, we can estimate the strength of interventions needed to maintain the bilingual subpopulation.
|•||A. Kandler, R. Unger and J. Steele (2010).|
Language shift, bilingualism and the future of Britain's Celtic languages. Phil. Trans. R. Soc. B. Vol 365:3855-3864 .
|•||W. Hoppitt, A. Kandler , J.R. Kendal and K.N. Laland (2010).|
The effect of task structure on diffusion dynamics: Implications for diffusion curve and network-based analyses. Learning & Behavior. Vol 38(3):243-251.
|•||Kandler, A (2009).|
Demography and language competition. Human Biology. Vol 81(2):181-210.