Cantonese Language Maintenance In Guangzhou Through The Lens Of Migrant Families’ Language Learning Investment

Thumbnail Image
Degree type
Doctor of Philosophy (PhD)
Graduate group
language investment
Language maintenance
language policy and planning
Other Languages, Societies, and Cultures
Urban, Community and Regional Planning
Grant number
Copyright date
Related resources
Liu, Yeting

The study is situated in Guangzhou. From the inception of China’s economic reform in 1980s, this city has been a southern hub for migrant workers from various socioeconomic backgrounds. As a historically Cantonese-speaking city, its linguistic ecology has been complicated by the national push for Putonghua (standardized Mandarin) and the influx of the migrant population who have different home languages and are assumed to opt for Putonghua as their lingua franca in Guangzhou. In the past decade, there were multiple incidents in which language advocates opposed and proposed local policies, fearing a language shift from Cantonese to Putonghua. In public discourse, the migrant population often gets blamed for that presumed shift. But have migrants learned Cantonese during their stay in Guangzhou? And are they motivated to learn? To find out how migrants contribute to this so-called Cantonese crisis, the researcher conducted a fifteen-month ethnographic study in one primary and one middle school in Guangzhou. These two public schools have received a large number of migrant students which accounted for half of their student populations. Informed by the language investment model (Darvin & Norton, 2015), a sociological approach to language learning motivation that takes into account the social positioning of language learners, and an ethnographic approach to language policy and planning that examines policy making and implementation at various levels accounting for both structure and agency (Hornberger & Johnson, 2007), the researcher administered questionnaires to 321 students and their parents (58% of whom were migrants without Guangzhou Hukou) from four grades, collecting demographic information and self-reported results on their language proficiencies, learning interests, and language use in different domains. This was followed up by participant observation in the schools, individual and group interviews with students, teachers and school administrators, linguistic autobiographies and family visits with focal students. Based on all the ethnographic data and questionnaire results, the researcher finds that the majority of migrant parents were active and successful learners of Cantonese while the children did not attain the same high level especially in speaking. This generational difference is intriguing as it disproves blanket statements claiming that the influx of migrant workers and their families caused the Cantonese crisis. In addition, the time of arrival and the socioeconomic resources adult migrants had when they first came to Guangzhou may have influenced their residential choice, social circles, and thus their investment in learning Cantonese. In particular, migrant parents in my study who came to this city with few resources but managed to make a living in service and sales were more likely to learn Cantonese as the language was related to their upward mobility. As for migrant students, unlike their parents’ more instrumental approach, their Cantonese language learning interest derived from affect, namely their socializing needs with peers and sense of belonging to the city of Guangzhou. So even though their self-reported language use suggested the majority of them were dominant in Putonghua, they are still quite invested in learning and improving their Cantonese. Through analyzing the institutional structure in schools and discursive practices of teachers and students and their underlying ideologies, the researcher also finds there are still ideological and implementational obstacles to overcome and avoid in order to pry open the spaces for either natural acquisition or formal instruction of Cantonese in school settings.

Nancy Hornberger
Yuko Butler
Date of degree
Date Range for Data Collection (Start Date)
Date Range for Data Collection (End Date)
Digital Object Identifier
Series name and number
Volume number
Issue number
Publisher DOI
Journal Issue
Recommended citation