National identity and multilingualism: a survey of Chinese Mongolian university students

ABSTRACT

In the field of multilingualism, limited research has explored the link between language proficiency and national identity, although the key role of language in the national identity (NI) formation has been acknowledged in other fields (e.g. sociology) over the past five decades. Aiming to narrow this gap, the present study examined Chinese Mongolian students’ NI and the influence from multilingualism (operationalised as self-rated proficiency in respectively Putonghua and English, their L2 and L3) and other sociobiographical variables (e.g. gender). NI was measured with a questionnaire item designed to gauge this construct that has been used in the latest waves of the Chinese Social Survey (CSS), a longitudinal nation-wide survey in China. The participants (N = 586) achieved a mean score of 4.26 (out of five) with the bootstrapped 95% confidence interval [4.20–4.32], indicating a high level of NI. Regression analyses showed that the two measures of multilingualism, attitudes respectively toward Putonghua, English, and trilingualism, and other sociobiographical variables influenced NI to different degrees. Calls for more comprehensive data concerning NI in multilingualism research and for more robust statistics (e.g. fuller use of effect size) were made. Policy implications vis-à-vis cultivating multilingualism among ethnic minorities were also discussed.

KEYWORDS:

• National identity
• proficiency in Putonghua
• proficiency in English
• attitudes toward Putonghua
• attitudes toward English
• attitudes toward trilingualism

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 In the present paper, multilingualism is defined as ‘at least partial mastery in a number of languages’ (Dewaele and Li 2013, p. 231), while it is acknowledged that ‘multilingualism can be operationalised in different ways, such as in terms of knowledge, ability, proficiency and frequency of usage’ (Cárdenas and Verkuyten 2021).

2 NI has been conceptualised differently in different disciplines including education, political sciences, and psychology. According to Tajfel’s (1982, 24) classic psychological definition, social identity refers to ‘that part of the individuals’ self-concept which derives from their knowledge of their membership of a social group (or groups) together with the value and emotional significance of that membership’. When the target social group is a nation, then the above quote represents a definition for NI. This NI definition, which is adopted by the present study, has its roots in Social Identity Theory (see Müller-Peters 1998; Guan and Guo 2019 for more).

3 In Wei and Hu’ (2019) effect size benchmark system, .005, .01, .02, and .09 respectively correspond to the small, typical (medium), large, and very large cut-offs for the effect size R²; the effect size r (roughly) has .07, .10, .14, and .30 as the small, typical, large, and very large benchmarks. This system has also been adopted in recent studies (e.g. Dewaele and Botes 2020) in the field of multilingualism. Regarding its applicability, this benchmark system ‘could potentially be applied to similar lines of survey research focusing upon psychological factors’(Wei and Hu 2019, p. 1216), although it was originally proposed to interpret the influence of sociobiographical variables on tolerance of ambiguity, a psychological ID similar to NI.

4 The Open-mindedness sub-scale comprises 13 items on the MPQ (long form) (van der Zee and van Oudenhoven 2000), as used in Korzilius et al. (2011). This number is reduced to eight on the MPQ (short form) (van der Zee et al. 2013), which has recently been used in recent studies (e.g. Dewaele and Botes 2020).

5 For instance, all of Han and Johnson’s (2021) Uyghur student interviewees receiving university education in Nanjing (the provincial capital of Jiangsu, outside of the Xinjiang Uyghur Autonomous Region) deemed it as their responsibility to study Chinese because they were Chinese citizens and regarded Chinese as a symbol of national identity; all of the interviewees in this qualitative study underscored not only their Uyghur identity and culture but also their Chinese-ness; one interviewee remarked, ‘As a Chinese, if you do not know your own language, and cannot write Chinese characters, what would foreigners think of you?’

6 The other less relevant studies conducted in the Chinese context are mostly conducted with a particular age group (e.g. Du and Fang 2013; Jiang 2015) or a specific ethnic minority group (e.g. Lin and Wu 2008; Wang 2008; Han and Cassels Johnson 2021; Sager 2021).

7 For the key statistics from the 2020 Census, see the Chinese Government website at http://www.gov.cn/guoqing/2021-05/13/content_5606149.htm.

8 For example, in the Guangxi Zhuang Autonomous Region, a major habitat for the Zhuang people (viz. China’s largest ethnic minority group), ‘all primary schools, either in the cities or in the rural areas, are required to teach English from primary three’ (Huang 2006, 54); graduates from secondary schools usually need to continue with learning English as a subject into the first two years in university (Wei et al. 2021). When discussing trilingual education in this autonomous region, Wu, Silver, and Zhang (2021) place the following three languages in juxtaposition: Zhuang, Chinese, and English. Indeed, ‘people in Guangxi now seem to be realising the importance of integrating into the global economy for development’ vis-à-vis English language learning (Huang 2006, 54), albeit their proficiency in English may be limited (cf. SGO 2006; Wei and Su 2015).

9 The 1984 Law on Regional National Autonomy defines an autonomous region as areas where ‘minority nationalities, under unified state leadership, practice regional autonomy in areas where they live in concentrated communities and set up organs of self-government’ (Han and Cassels Johnson 2021). In other words, the IMAR, like other autonomous regions, has the authority to establish its own education system, decide upon the medium of instructions, and determine which foreign language to teach in school. During pre-tertiary education, most schools provide some form of trilingual education, with Chinese (usually from Primary One) and English (at least from Primary Three) being taught as school subjects; at some schools with sufficient resources, Chinese is used as a teaching medium not only for the subject of Chinese but also for other non-language subjects (Wei et al. 2021). During tertiary education, Mongolian students have the right to receive their education through two language systems, Mongolian or Chinese (Dong et al. 2015), where they normally need to continue with English learning in their first two university years; although unlike Putonghua, English as a foreign language does not enjoy official language status in China, nowadays many ethnic minority students seem ‘fully aware of the importance of English’ (Ma and Luo 2017, 67).

10 For instance, Guan and Guo (2019) conceptualised NI as a multidimensional construct comprising as many as nine specific dimensions including self-categorisation and social embeddedness. In contrast, Pöllmann (2008) distinguished between civic and ethnic national identities and used only two items to measure the bidimensional construct of NI; the respondents, i.e. 281 Berlin and London state secondary school head teachers, were given the neutral instructions that ‘Some people say that the following criteria are important for being truly British/German. Others say they are not important’ and invited to indicate their perceived importance respectively of ‘to have British/German citizenship’ and ‘to have British/German ancestry’ on a four-point Likert scale. In the Chinese context, Zhang and Xu (2013) similarly conceptualised NI as bidimensional and developed over a dozen items on a four-point Likert scale when surveying 258 Mongolian university students from one Chinese city. Interestingly, Li and Liu (2018) used as few as five items to measure NI, which was assumed to be unidimensional (Cronbach’s alpha = .724).

11 Li and Liu (2018) (in Note 10) utilised data from the 2013 wave of the Chinese Social Survey (CSS) based on a nationally representative sample in China; most recently, Wei et al. (2022) drew on more recent data from the same source (viz. the 2015 CSS), which removed four out of the above-mentioned five items and kept only one item for the measurement of NI, to investigate the link between bilingualism and NI. The single-item measurement practice (e.g. Xia 2019) is also adopted by non-China based researchers; for example, Cárdenas and Verkuyten (2021) in the Netherlands operationalised NI as the answer to one question (‘How strongly do you identify as a Dutch person?’).

12 Results from bivariate analyses tend to generate inflated effect sizes (Wei, Reynolds, Kong, and Liu 2022). Even when more than one focal variables (e.g. two) are used as independent variables in regression analyses, the risks for generating inflated effect sizes largely remain (cf. Wang et al. 2022). Hence including a few more variables (e.g. gender and other socio-biographical factors) together with the focal variables as predictors into regression analyses will facilitate a better understanding of the influence of focal variables on the dependent variable.

13 While some endeavour of different stakeholder groups (e.g. policy makers and students) to negotiate for a more balanced and equitable language education is documented (Adamson and Feng 2009; Feng and Adamson 2015; Wu, Silver, and Hu 2022), the outcomes of such endeavour remain unclear and the agency of minority groups should receive (continued) support to enable their empowerment and the cultivation of their multilingualism.

14 Here, a bigger group in terms of linguistic vitality is loosely defined as an ethnic minority group whose language has a more functional writing system. Based on the history of the writing system development of their languages, China’s minorities can be categorised into three groups. Group 1 consists of the Korean, Kazak, Mongolian, Tibetan and Uygur, which had functional writing systems broadly used before 1949. Group 2 includes the Dai, Jingpo, Kisu, Lahu, Miao, Naxi, Va and Yi, which had functional writing systems of limited usage before 1949. The remaining 42 ethnic minority groups comprise Group 3, which had no fully functional writing systems prior to 1949 (Zhou 2000; Wei, Jiang, and Kong 2021).

Additional information

Funding

This work was supported by Chinese Society for Multilingualism and Multilingual Education affiliated to the International Association of Multilingual Education; Xi'an Jiaotong-Liverpool University [grant number REF-19-02-01].