Spatial Distribution of Language Use between and within Generations of Multilinguals in a Rural African Region

Authors: Ling Bian*, University at Buffalo, Penghang Liu*, University at Buffalo, Pierpaolo Di Carlo, University at Buffalo, Jeff Good, University at Buffalo, Yujia Pan, University at Buffalo
Topics: Geographic Information Science and Systems, Africa
Keywords: Linguistic distribution, Spatial networks, Rural Africa
Session Type: Poster
Day: 4/6/2019
Start / End Time: 8:00 AM / 9:40 AM
Room: Lincoln 2, Marriott, Exhibition Level
Presentation File: No File Uploaded

The Lower Fungom region of Northwest Cameroon is known for its high language density. For example, in an area of 200 sq km, there are more than eight distinct languages associated with 13 villages. Also, most people normally speak three or more local languages plus Cameroonian Pidgin English and some English. To understand such multilingualism and to aid in the preservation of language diversity, this study aims to obtain baseline understanding of the spatial distribution of the five most used languages between and within two generations of multilingual villagers.

Individual-level language use is represented by a feature vector model. A vector consists of 26 feature values, expressed as the number of villagers who speak the language(s) in 26 villages of the wider region. One feature vector is constructed for 1) each language and each combination of two to five languages and 2) for each of three groups of villagers, the ego, father, and mother. Five distance metrics are used to analyze the similarity in language use in the region between and within generations.

From a spatial perspective, the top five languages are dominant in their designated and immediately adjacent villages. From a temporal perspective, between-generation differences are considerably greater than within-generation differences. This trend may have ramifications for the spatial distribution of future generations of multilinguals in this rural area of central Africa.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login