The Politics of Language Contact in the Himalaya

This book brings together linguistic theory and empirical studies addressing human rights, multilingual education, language ecology and endangered languages. It is essential reading for students, practitioners, language activists and scholars working on language planning, multilingual education, endangered languages and language politics. This is indeed an interdisciplinary book that is testimony to why lesser-known languages matter in the Himalaya and beyond. —Prof. Nirmal Man Tuladhar, Chair, Social Science Baha


Bendi Tso and Mark Turin
The People's Republic of China (henceforth, PRC) is firmly committed to a language policy of promoting the Chinese language among ethnic minority groups and encouraging ethnic minorities to construct a Chinese national identity through linguistic and cultural assimilation.
Running parallel to such assimilationist policies are standardization projects that recognize China's heterogeneity, and preserve and promote ethnic minority languages that are sponsored by the PRC government and maintained by ethnic minorities. The state has succeeded in both of these ideological and strategic endeavors. The Chinese language and associated cultural norms have been widely adopted and ethnic minority languages have effectively been standardized. However, as a collateral byproduct of these processes, non-standard dialects of ethnic minority languages have been marginalized and begun to disappear. This chapter examines official language policy as it is played out in Chone County, located in the Kanlho Tibetan Autonomous Prefecture (TAP) of Gansu Province, and explores how state policy interacts with locally-held language ideologies among Chone Tibetans. Using the concept of hegemony, we argue that China's language policy and Chone Tibetans' willingness to adopt new languages serve to generate both coercion and consent. These two forces work together to contribute to a three-tiered linguistic hierarchy that has emerged from an intensification of language contact since the 1900s, if not earlier. This three-tier linguistic hierarchy, with Chinese and Amdo Tibetan as the dominant languages -albeit with Chinese firmly situated above Amdo Tibetan -and Chone Tibetan as the subordinate language, has accelerated the loss of Chone Tibetan through the dual oppression from two linguistic hegemonies that have exerted complex pressures on the speech community of Chone.
In this chapter, we first present the theoretical framework of our study and discuss how forces of coercion and consent work together to secure and maintain the dual linguistic hegemonies. We then proceed to sketch out the context of our research and our methodology. Next, we illustrate how literacy campaigns and educational policies in Chone function to coercively establish the linguistic hierarchy. We then discuss the apparent willingness of Chone Tibetans to internalize this linguistic hierarchy and to adopt new languages, and how this helps to generate a sense of consent that in turn legitimizes the hierarchically superior positions enjoyed by Chinese and Amdo Tibetan. Our goal, throughout this chapter, is to contribute to an emerging and critical conversation about linguistic diversity within cultural Tibet.

The Shape of Linguistic Hegemony: Coercion and Consent
Building on Woolard's (1985) and Wiley's (1996) notions of hegemony, for the purpose of this study we define linguistic hegemony as the position and power that the dominant language enjoys in legitimating and reproducing prevailing ideologies and practices. The hegemony to which we refer here is linguistic and cultural rather than political. The adoption of a particular language by a group of speakers is never random nor is it accidental (Carlucci 2013;Ives 2004;Suarez 2002;Wiley 1996). Instead, it is assisted, shaped and promoted by coercive forces. And the consent displayed by speakers in turn justifies and secures the existence and practice of coercion. Consent and coercion work together in powerful and socially effective ways to establish and secure the superior position enjoyed by the dominant language or languages.
To better understand how linguistic hegemony is created and reproduced in the context of Chone, we examine two interactive aspects of hegemony: coercion and consent. The first point of analysis is located within the powerful nature of hegemony itself and explores how state institutions and policies enable and maintain linguistic domination. Social institutions, including but not limited to law and language policy, shed light on the politics of language by producing expectations of 'standard' language and by taming diverse language ideologies held by the people so that only a standard language appears to be natural and proper (Eriksen 1992;Wiley 1996). Through powerful social institutions, the linguistic hierarchy is established. Language norms are shaped and then take hold. Furthermore, linguistic hierarchy is produced and reinforced by social institutions that lead to speakers of minority languages and varieties eventually accepting an ideology that defines their heritage language as inferior, ultimately leading to its abandonment (Eriksen 1992).
Yet, at the same time, we acknowledge that 'authority and hegemony cannot be mechanically read out from institutional dominance' (Woolard 1985: 743) since the successful normalization of a linguistic hegemony relies on the extent to which 'linguistic minorities will believe in and participate in the subjugation of the minority language to the dominant, to the point where just the dominant language remains' (Suarez 2002: 514). In other words, institutions alone cannot fully sustain the power of a dominant language. The consent of speakers of minority languages, and their willingness to adopt a dominant language, is another key aspect of hegemony, and one that grants the coercive force important legitimacy. In contrast to coercion, consent is achieved through the apparently willing participation of minority language speakers in promoting, accepting and appreciating the dominant language (Ives 2004;Suarez 2002;Wiley 1996). In the context of consent, minority language speakers' adoption of a dominant language is not an 'accommodative behavior' (Woolard 1985: 741) that conceals their true and deep feelings of value towards their heritage language and thus accommodates to the coercive force. Rather, their internalized language ideology becomes part of the 'collaborative consciousness' (Woolard 1985: 741) and is naturalized, thus working to shape their daily language practice and reproduce the linguistic and cultural hegemony. It is, then, through their language ideologies that minority language speakers participate in the social reproduction of consent, and the cultural and linguistic domination over them is justified, secured and sustained.
To summarize, the superior position enjoyed by a dominant language cannot be established without a combination of institutional coercion and the consent of minority language speakers. With the support of institutional powers, a dominant language becomes the yardstick for establishing a linguistic hierarchy, which in turn makes minority language speakers accord it high value (Turin 2018). Through the participation of minority language speakers, language ideologies that offer the dominant language prestige, pride of place and authority are given legitimacy. This iterative process secures and reinforces the position of the dominant language.

Situating Chone County in Time and Place
Chone County is located on the Sino-Tibetan border. It was incorporated into Kanlho Tibetan Autonomous Prefecture (henceforth, Kanlho TAP) of Gansu Province, China, in 1953. Ethnolinguistically speaking, Chone is located in Amdo, one of three cultural-linguistic regions of cultural Tibet.
According to Choephel (1978), Chone was an ancient border outpost of the Tibetan empire. The earliest definitive evidence showing the settlement of Tibetans in this area dates to the seventh century, when Songtsen Gampo -the famed King of Tibet -sent his troops to Chone. Chone is also home to several other ethnic groups, including but not limited to Chinese, Hui Muslims, Mongols and Monguors. Chone County is closely interlocked with Batse County of Kanlho TAP, the majority of whose residents are Hui Muslims and Chinese. In broad strokes, Chone shares its northern, western, and southern borders with Tibetan communities and its eastern border with Chinese communities.
As a multi-ethnic border zone, the political and linguistic history of Chone is rich and complex. Chone is marked by 600 years of governance by a Tibetan chieftain family who maintained religious connections with Lhasa in Central Tibet while winning political recognition from Chinese courts (Rock 1933;Yang 1990). Also noteworthy is the massive Chinese immigration into Chone that began in the early twentieth century due to 5. Speaking Chone,Speaking 'Shallow' Fig. 5.1. Map of Chone County as situated within Amdo, within China. Adapted from Kolås and Thowsen (2005: 2), CC BY-NC-ND.
the Hui Muslim revolt in 1928 in He Zhou -which came to be referred to as the He-Huang Incident -located in what is today the Linxia Hui Autonomous Prefecture, 1 adjacent to Kanlho TAP (Chone Gazetteer Compilation Committee 1994: 166). Since the 1950s, the PRC's national language policy has promoted Chinese in Chone. Also, since the late 1980s, the Amdo Tibetan standardization movement -encouraged by the Chinese state to preserve and develop the Tibetan language, and later instigated by Tibetan political and cultural elites to safeguard the survival of the Tibetan language and assert Tibetans' Tibetanness -has further complicated the already complex ethnolinguistic landscape of Chone. Broadly speaking, there are three identifiable Tibetan linguistic groups within Chone County, according to their geographical location and complex histories of migration: 1) the group of nomads who live in the western and northern pastoral regions of Chone and whose spoken language is close to Amdo Tibetan; 2) the group of farmers who live in the eastern enclave of Chone and whose spoken language shares linguistic feature with both Kham and Lhasa Tibetan; and 3) riverside Tibetans who live along the Luchu River and its tributaries, and whose spoken language is closer to Lhasa Tibetan (Chone Gazetteer Compilation Committee 1994: 164, 699). Our research site is located in the riverside Tibetan community. Within the official PRC discourse, 'riverside Tibetans' are recognized to be Chone Tibetans and the ethnonym has been accepted by locals. For the purpose of this study, then, we refer to the riverside Tibetan community as 'Chone Tibetans' and to their spoken tongue as the Chone Tibetan language. The Chone Tibetan language is only intelligible to speakers living alongside the Luchu River and its tributaries, making its linguistic ecology somewhat unfavorable. By this we mean that Chone Tibetans live in a traditional Amdo Tibetan-speaking area, but speak a form of Tibetan that is, if anything, closer to Lhasa Tibetan. 2 Due to their frequent, intensive and We are aware that linguists critically challenge the notion that Amdo, Lhasa and Kham are three dialects of the 'same' Tibetan language, and that they highlight the linguistic diversity and complexity that exists within these three dialects (Suzuki prolonged contact with the Chinese language over time, Chone Tibetans have increasingly been exposed to strong assimilationist forces from mainstream Chinese culture, and have adopted many Chinese loan words into their language. Our principle research site was Yarru village, a settlement located along the northern side of the Luchu River. According to oral histories passed down through the generations and augmented by interviews with Chone Tibetans in Yarru village, Tibetans from Central Tibet settled in the area thanks to its hospitable and fertile environment. There are currently sixty-four households in Yarru village, of which at least twelve are Chinese. The Chinese settlement can be accounted for in two ways: some Chinese were hired by Chone Tibetans as farm laborers in the early twentieth century, 3 while others were refugees who suffered during the Hui Muslim revolt in the 1920s and fled to Chone. The cumulative result of this Chinese immigration is that, since the early twentieth century at least, Chone Tibetans have been gradually exposed to both written and spoken Chinese language and culture. According to the interviews with Chone Tibetans in this village, during the earlier period of Chinese settlement, language shift did not occur because Chone Tibetans significantly outnumbered ethnic Chinese. Later, when a series of political and cultural events provided strong incentives for Chone Tibetans to incorporate Chinese loanwords into their language and to learn to speak, read and write Chinese, this triggered widespread language shift which further promoted the belief that the Chone language was a 'low-status' Tibetan language. As a result of these interconnected factors, many Chone Tibetans are now experiencing a growing sense of what may be described as linguistic and cultural anxiety and alienation. 2016; Tournadre 2014). However, we use these terms in order to emphasize the cultural hegemony associated with these languages instead of their linguistic distinctiveness per se. Meanwhile, there is a disparity between speakers' locally conceptualized linguistic categorization and those defined by linguists. Based on the data collected from interviews, local Chone Tibetans tend to believe that their language is similar to Lhasa Tibetan. 3 This labor force later developed into the Small Households, which refers to the shelters that some long-term laborers and refugees built at the outskirts of Tibetan villages along the Luchu River (Chone Historical Accounts Compilation Committee 1984: 10; Lhamo, Interview, 23 May, 2017).

Research Methods and Subject Position
This research is based on interviews and observations conducted with community members between May and August in Yarru village in 2013, a series of follow-up interviews with Chone Tibetans in Yarru village between October and December 2016, and ethnographic fieldwork conducted between mid-May and late June 2017 in Yarru village and the neighboring villages. Our research questions focused on language practice, shift and ideology among Chone Tibetans in the context of China's wider language policy. In the field, the lead author's identity as an ethnic Chone Tibetan with distant relatives living in Yarru village facilitated her acceptance within the community. Participant observation served as her primary research method, which -when combined with familiar contact and continual encounters with community members -opened up many spaces for designing and honing complex research questions. Semistructured interviews with county, township and village officials helped her to explore the role played by institutions and bureaucrats in shaping Chone Tibetan language ideologies. We also used semistructured interviews to explore Chone Tibetans' language proficiency and attitudes towards their heritage language, as well as towards the more dominant Amdo Tibetan and Chinese languages. Semi-structured interviews with parents, children and village school teachers helped us understand their experiences of language acquisition, while life story interviews with elder Chone Tibetans were particularly useful in documenting shifting experiences of and changing attitudes towards language acquisition. All of these methodological techniques were used to assess the processes of consent and coercion that are integral to the formation and normalization of dual linguistic hegemonies.
Interviews were conducted in the local Chinese dialect, which is known as the Taozhou Chinese dialect. It belongs to the Northwest subdialect of the Northern dialect of Chinese (Wang 2013), and it has now become the lingua franca of Chone and Batse Counties. It was necessary to use the local Chinese dialect to conduct interviews because the majority of Chone Tibetans under the age of forty are now monolingual in Chinese, and generally speak the local Chinese dialect amongst themselves. Second, the lead author's imperfect proficiency in Tibetan limited her ability to conduct interviews in her ancestral language. In this chapter, we use pseudonyms for research partners and locations to mitigate any unwanted and unwelcome repercussions.

Coercion as an Aspect of Linguistic Hegemony
The adoption of a dominant language by minority language speakers and the establishment of a linguistic hierarchy are not processes that occur in a vacuum. Instead, these actions are supported by the policy contexts in which they occur. The national language policy of China plays a central role in constructing and maintaining the linguistic hierarchy and triggering language shift from ethnic minority languages to Chinese. To date, there have been two dominant discourses within the PRC's language policy. First has been the promotion of Chinese in almost all domains, including education, the workplace, and the media, at both national and community levels (Spolsky 2014;Rohsenow 2004;Wang and Yuan 2013). Second, the state has offered incentives and some resources to encourage the preservation, standardization and development of minority languages (Harrell 1993;Zhou 2004;Li and Huang 2004;Li and He 2014). Laws and regulations have been implemented by both central and local governments that enshrine and guarantee minorities' rights to use and develop their own heritage languages. Within the Kanlho TAP, the rights to use, learn, preserve, standardize, and develop Tibetan are enshrined in the Autonomy Statute of Kanlho Tibetan Autonomous Prefecture of Gansu Province, which was passed in 1989 and amended in 2010 (Articles 24, 28, 60, 63, 64, and 74).
Even though these two distinct discourses are emphasized within the PRC's language policy, they do not guarantee that minority languages have a linguistic position equal to Chinese or that minority languages can access the same range of discursive and financial resources that Chinese does (Zhou 2012). The government -through the state's language policies and associated ideology -decides which language can be used in which context, with the result that languages are accessed to differing degrees, forming a language order or hierarchy (Zhou 2012). As Arienne Dwyer (1998) observes, a five-tiered linguistic hierarchy has been established in the PRC. This hierarchy positions Mandarin Chinese at the top and situates standardized minority languages over 'primary minority languages, secondary minority languages and unrecognized languages' (Dwyer 1998: 71-72). The niche that a language occupies within this five-tiered linguistic hierarchy is highly dependent upon the prestige and political influence of the language, and the number of speakers (Dwyer 1998). The state, in turn, offers various levels of support to different languages depending on their position along and within this hierarchy (Dwyer 2005). In the case of Tibetan, the three main vernaculars (Lhasa, Kham and Amdo Tibetan) have become the most prestigious Tibetan language varieties in their respective regions. Amdo Tibetan -the dominant language in Amdo region (Roche 2017) -has been promoted over local Tibetan varieties because its pronunciation is closer to literary Tibetan (Prins 2002). In addition, the nomadic pronunciation of Amdo Tibetan adds extra prestige to the language since nomadic life is often imagined as the key aspect of Tibetan identity (Prins 2002).
In this section, by examining literacy campaigns and educational policies implemented in Yarru village from the 1950s onwards, we argue that the PRC's coercive language policies are powerful forces for the establishment and maintenance of hegemonies. These influences have prompted language shift from Chone Tibetan to Chinese, and have helped to establish Amdo Tibetan as the effective standard Tibetan language for this region.

The Role of Literacy Campaigns
Following the foundation of the PRC, on account of the powerful symbolic connection between literacy and the imagining of the nationstate, the government set out to address illiteracy in Chinese as well as in other regionally-dominant ethnic minority languages through national and local policies (Spolsky 2014;Zhou 2005). At the national level, ethnic communities were encouraged to improve their literacy in ethnic minority languages in the early period of the PRC, especially during the 1950s (Zhou 2005). Article 6 of the Regulations on the Eradication of Illiteracy, which was passed in 1988 and which was later amended in 1993, states that the medium of instruction for literacy campaigns should ideally be Mandarin Chinese, but that in ethnic minority areas, either the ethnic minority language or the language most commonly used by all of the ethnic minorities in a community could be the medium for literacy education. At the Prefectural level, Article 9 of the Regulations on Tibetan Language Use in Kanlho TAP (1996) stated that the campaign to improve Tibetan literacy among Tibetans should be systematically implemented. At the County level in Chone, Tibetan learning materials were distributed in 2005 in certain villages where the Tibetan language has been used as the medium of education to combat Tibetan illiteracy (Chone Education Bureau 2011: 82). In Yarru village and other villages where Chone Tibetans reside, however, the seven-decade-long literacy campaign was focused on increasing Chone Tibetans' literacy in Chinese rather than Tibetan. This has resulted not only in Chone Tibetans learning to read and write in Chinese, but has also accorded high value to Chinese since 'literacy is a primary vehicle for maintaining and transmitting cultural values and beliefs' (Zhou 2005: 117-18).
In Yarru village, the shape taken by literacy campaigns has changed over time, but their purpose -namely exposing Chone Tibetans to the Chinese language -has remained the same for seven decades. In the early years, from 1953 to the mid 1960s, literacy classes were the major expressions of literacy campaigns (Chone Education Bureau 2011: 20; Chone Gazetteer Compilation Committee 1994: 550). These literacy classes were not only classroom-based, but were also held in informal locations throughout the village. Teachers, students and cadres who were literate in Chinese participated in teaching Chinese characters to illiterate Chone Tibetans. Lhamo, a grandmother now in her mideighties, recalled that in the 1960s, after a day of hard work on the farm, Chone Tibetans would gather together to learn Chinese. Lhamo spoke of how local Chinese cadres and Tibetan students who had received education in the Chinese language taught basic vocabulary such as numbers and commodities, introducing her to new terms. Even though the classes were held only intermittently, Lhamo reported that after several years of learning, she could write her name in Chinese and had learned basic vocabulary. Along with acquiring rudimentary Chinese, Lhamo was also taught that old Tibetan kinship terms should be replaced with Chinese ones: 'Ama (mother in Chone Tibetan) should be muqin (mother in Chinese); Ape (father in Chone Tibetan) should be fuqin (father in Chinese)'. Through the erosion and belittling of traditional Tibetan terminology -even in the domain of kinship -the supremacy of the Chinese language and cultural hegemony with which it was associated began to be established.
During the Cultural Revolution (1966)(1967)(1968)(1969)(1970)(1971)(1972)(1973)(1974)(1975)(1976), literacy campaigns were escalated. Alongside literacy classes, villagers were expected to memorize Quotations from Chairman Mao as a foundational Chinese language-learning resource. Chinese language learning was evaluated by reading quotes out loud from blackboards, which were even mounted outside the communal dining hall of Yarru village and the ferry pier along the Luchu River. Bendi Gyal, a grandfather now in his mid-seventies, recalled that two students from a local school usually stood beside the blackboard and would ask villagers to read the quotes aloud. If they were unable to read the quotes correctly, they would not be permitted to eat or take the ferry, and students would offer instruction in the quotes until villagers could read them. When the policy was tightened further, blackboards were mounted at the entrance of villages. Every time villagers passed by the gateway, they were required to read the quotes on the blackboards. Although we cannot say to what extent this literacy campaign helped Chone Tibetans acquire Chinese, such practices certainly changed Chone Tibetans' attitudes towards the Chinese language in which Chinese was positioned as a resource that could bolster status and garner prestige. As Lhamo said, 'when we were young, those who could speak and write Chinese were [thought to be] the talented ones.' After 1976, each household received a radio from the government. News from China National Radio and Chone County Radio were broadcast in Mandarin Chinese at 7pm every day, introducing Mandarin Chinese to the community in a more systematic way. Since the 1990s, literacy campaigns in Yarru village have become increasingly diverse and multimodal, including -but not limited to -meetings, songs, TV, blackboards, posters, and job trainings. In 2005, the literacy campaign strategy gradually shifted its focus from teaching Chinese characters to receiving skills training in the Chinese medium (Chone Education Bureau 2011: 83). Literate Chone Tibetans were told that they would become more competitive in the labor market through reinforcing the Chinese characters that they had learned. Courses introducing skills in vegetable cultivation, animal husbandry, construction, noodle making and using sewing machines have been offered to people in Yarru village through the medium of Chinese, all on the understanding that this will make them more competitive in the labor force.
Even though specific statistics on changes in Chone Tibetan literacy rates as a result of this seven-decades-long campaign are still lacking, 4 overall literacy rate changes in Chone County are available. While the illiteracy rate was above 90% before the foundation of the PRC in Chone (Chone Gazetteer Compilation Committee 1994: 165), by the end of 2010, the illiteracy rate in Chone had dropped to 3.14% (Chone Education Bureau 2011: 84). Through the waves of successive literacy campaigns, Chone Tibetans now not only know how to speak and write Chinese, but have also internalized the prevailing linguistic order. The strong and sustained position taken by the government has helped to establish the linguistic domination of Chinese.

Changing Educational Policies
While the national literacy campaign outlined above introduced the Chinese language to Chone Tibetans and helped to establish a linguistic hierarchy with Chinese as the superior language, educational policies worked to further complicate the linguistic hierarchy by having Chone Tibetans learn both Chinese and Amdo Tibetan in a structured and formal way, speeding up the loss of the Chone Tibetan language. As widely documented (cf. Aronowitz 2002;Macedo, Dendrinos and Gounari 2003;Wiley 1996), schools are important sites for a dominant language and its associated culture to be asserted, justified and reinforced. Schools are also important locations where people's consent is acquired and linguistic hegemony is reproduced. In China, the stated aim of bilingual education is to introduce the Chinese language to ethnic minorities and have ethnic minorities develop Chinese language proficiency in order to better assimilate them into Chinese society (Dwyer 2005). At the same time, as an intended or unintended byproduct, bilingual education helps to establish the superior and authoritative position of standardized minority languages over local varieties, as the standardized local language is chosen as the medium of instruction in school settings. In the case of Yarru village, educational policy has legitimized Chinese and elevated Amdo Tibetan as the regional standard, which in turn has encouraged Chone Tibetans to view their heritage language as inferior and less formal. The dual hegemonies engendered by this bilingual education policy have contributed to the rapid loss of Chone Tibetan and to a growing sense of linguistic anxiety among its former speakers.
The primary school in Yarru village was founded in the 1950s at the same time as the establishment of the People's Commune; because of political campaigns, classes were not regularly offered until the mid-1970s. Chone Tibetans recall that since the founding of the school, the medium of instruction has always been Chinese, and that Tibetan classes were not offered until 1987. The only opportunity for Chone Tibetans in Yarru village to learn a Tibetan language in school before 1987 was to enroll in the Town Center Middle School where both Tibetan and English were taught as school subjects. Pema, aged forty-seven, was educated in the Town Center Middle school, and confirmed that standardized Amdo Tibetan was taught there twice a week. When asked about his experience of learning Amdo Tibetan in a school setting, he answered, 'our [Chone Tibetan] language is different from the spoken language taught in school, which is taught by teachers from regular school.' From Pema's point of view, the major difference between Amdo Tibetan and Chone Tibetan was formality. He further added, 'Ironically, the Tibetan language class was later cancelled in the second year while the English class was never cancelled in all of my three years. Just as in English, we are dabblers in Tibetan.' Pema is an ethnic Tibetan whose native language is Chone Tibetan. However, the standard form of the language taught in school has changed his understanding of the Tibetan language. Having described himself as a dabbler in Tibetan, he further perceives Chone Tibetan to be linguistically deficient, inauthentic, and subordinate when compared to Amdo Tibetan.
In 1987, when the Tibetan language was finally offered for instruction in Yarru village twice a week under Model Two of China's bilingual education policy, 5 it was the standardized Amdo Tibetan language 5 Currently, two bilingual models have been implemented in the Tibetan areas of China. The first, known as 'Model One', refers to an approach where the medium of instruction is Tibetan and all textbooks are written in Tibetan, except for those used in Chinese and English classes. The second, known as 'Model Two', refers to an approach where the medium of instruction is Chinese and all textbooks are written in Chinese, except for those used in Tibetan and English classes that was selected to be taught. The first Tibetan language teacher, Dorje Men, is an ethnic Chone Tibetan born in Yarru village. She was stationed in Yarru village immediately after graduating from Kanlho Minorities' School in 1987 where she first learned to write Tibetan using standardized Amdo Tibetan over the course of three years. She recalled that back then, most of her students could speak Chone Tibetan but they did not know how to read or write in any Tibetan language. She therefore corrected their pronunciation and taught the Tibetan script in standardized Amdo Tibetan form, so that both standardized vocabulary and pronunciation would be imparted to the students. Through such approaches, a linguistic hierarchy was established within schools with the Chinese language as dominant and Amdo Tibetan as secondary and subordinate, and this hierarchy was maintained through the medium of instruction. However, two years later, Dorje Men was transferred to a primary school in the Chone County seat to teach math and music. When villagers were unhappy with her departure and asked the principal of the Town Center School to assign a new Tibetan teacher to the village school, their request was not honored due to a shortage of recognized Tibetan language teachers. Later, several Tibetan language teachers were stationed in Yarru village school but none of them stayed long. As a result, the educational model of Yarru village school was finally switched from Model Two to Model Three, and Tibetan classes were permanently canceled (see n.5 for more information on the different educational 'Models'). This situation lasted until the Yarru village primary school was permanently shuttered in 2013. (Yang 2017). In addition to these two bilingual models, 'Model Three' refers to an approach where the medium of instruction is Chinese, and all textbooks are written in Chinese except for the ones used in English class. 'Model Three' does not include the provision of a Tibetan class. These three models were implemented based on both geographical location and the modes of production on which people have relied (Personal communication with cadres). 'Model One' was implemented in the pastoral region of Chone. 'Model Two' was implemented in farming-pastoral regions, where most Chone Tibetans, including Tibetans in Yarru village and some Chinese, reside. Nowadays, apart from one primary school in a village where Chone Tibetans reside that still follows 'Model Two', the rest of the schools that Chone Tibetans attend, including the primary school in Yarru village, have gradually shifted to 'Model Three' due to the lack of Tibetan language teachers and at the insistence of parents. Such a policy has its own limitations since Tibetan classes were offered based on the modes of production of the community rather than students' ethnicities, thereby arguably not meeting the real needs of Tibetans who wish to learn their Tibetan language.
From 2013 to the present, children in Yarru village either attend the Town Center Boarding School or Chinese schools in the Chone County seat. No matter which school they choose, the medium of instruction and all of their textbooks are in Chinese, and Tibetan has increasingly become a foreign and distant language for them. At the same time, Mandarin Chinese is being robustly promoted in school. Teenagers have described a Language Commission Office that was set up in the Chinese high school in the Chone County seat, with two teachers in the office taking charge of promoting Mandarin Chinese. Also, because most teachers working in the Chone High School were not local and spoke only Mandarin Chinese, the local Chinese dialect was not widely tolerated in school and teachers strongly encouraged teenagers to use Mandarin Chinese for all communication. The same process occurred in the Town Center Boarding School. Slogans such as 'please speak Mandarin Chinese, please write standardized Chinese characters' were displayed and conveyed the prevailing belief that only Mandarin Chinese in standardized Chinese characters was a legitimate language within the school setting. Children's extensive exposure to Chinese in school made parents believe that it was the children's Chinese learning that triggered language shift within the family. During fieldwork, the principal author noted that the sentiment that 'ever since the children were sent to school, the transmission of the (Chone) Tibetan language is over' was heard time and again.
This review of seven decades of educational policy changes in Yarru village illustrates how policies have shifted back and forth between promoting Chinese, on the one hand, and supporting Chone Tibetans to learn Amdo Tibetan on the other. This oscillation can be explained by understanding how China's language policy is tightly connected to its policy relating to ethnic minorities and nation-state building, both of which have been subjected to changes in response to state-ethnic minority relations over the years (Zhou 2003). Dwyer categorizes such dramatic policy pivots as 'ideologically and pragmatically chaotic ' (2005: 22). A consequence of such frequent changes is the creation not only of 'generation gaps' in terms of language learning and literacy development, but also 'gaps in historical knowledge' (Beckett and Postiglione 2012:14). In the case of the Chone Tibetans of Yarru village, frequent educational policy changes not only deprived them of their right to learn the Tibetan language, but also resulted in most generations being fully exposed to Chinese language and culture, thereby speeding up the production and reproduction of Chinese linguistic and cultural hegemony and the assimilation of Tibetans.
In today's Chone, following the cancelation of Tibetan classes, Chinese has become the single most prioritized language in the school setting. However, this has not resulted in Amdo Tibetan losing its considerable influence in Chone County. Chone Tibetans are still under the linguistic and cultural influence of the Amdo Tibetan language, an influence which at times can feel oppressive. Recently, guided by the Regulations on Tibetan Language Use in Kanlho TAP of Gansu Province (1996), the Article 24 of the Autonomy Statute of Kanlho TAP of Gansu Province (2010), and the Implementation Rules for the Regulations on Tibetan Language Use in Kanlho TAP of Gansu Province (2006), Tibetan language and its use, development and promotion within Kanlho TAP has been strongly encouraged in both letter and spirit. In Chone, the Tibetan Language Committee was formed in 2010, and consists of five staff of whom four majored in Tibetan Language and Literature. The Committee is tasked with implementing six goals, one of which is to promote Tibetan language standardization and the use of Tibetan in Chone. Even though it is nowhere explicitly stipulated which form or variety of Tibetan should be regarded as the standard language in Chone, the lead author of this chapter was informed that the unwritten rule is that it be Amdo Tibetan. Consequently, Amdo Tibetan is widely used in formal linguistic contexts in Chone when Tibetan is called for. For example, shop signs and road signs are written in Amdo Tibetan; the Chone Television Station broadcasts in Amdo Tibetan; and the medium of instruction in Tibetan class at the Tibetan High School of Chone County is Amdo Tibetan, with accommodations made for different teachers' accents. Supported by the state, then, Amdo Tibetan is now encroaching into every domain in which Chone Tibetan was once used, and is even expanding into new domains.
As is clear from the above description, educational language policies implemented by the central and local governments are also hegemonic projects. These policies not only inscribe and reinforce a linguistic hierarchy -which gives priority to Chinese and Amdo Tibetan through the medium of instruction, textbooks and public signage and media -but also serve to create a sense of language deficit and linguistic alienation, whereby Tibetans in Yarru village lose the opportunity to learn their own heritage language.

The Role of Consent in Shaping Linguistic Hegemony
As noted by Peter Ives, 'language is spread predominantly not by government or state coercion, military or police action, but by speakers accepting the prestige and utility of new languages, phrases or terms ' (2004: 7). By this we understand that the willingness of minority language speakers to adopt a dominant language leads to the reproduction of a linguistic hegemony, thereby offering the dominant language further legitimacy. The rationale behind people's attitudes and choices about the languages they adopt is socio-political, economic and cultural (Kroskrity 2000). In the context of Chone, we argue that Chone Tibetans have internalized the dominant belief that Chinese is a more modern and useful language than their variety of Tibetan, leading to pronounced language shift. At the same time, Chone Tibetans have also internalized a related if different language ideology: that Amdo Tibetan should be thought of and promoted as the standardized Tibetan language in formal and public contexts. The internalization of a three-tiered language hierarchy with Chone Tibetan at the bottom of the stack has resulted in Chone Tibetans voluntarily participating in the legitimization and reproduction of dual linguistic hegemonies.

Tracking Language Shift from Chone Tibetan to Chinese
Supported by China's language policy and nation-state building, Chinese is the language associated with notions of science and modernity (Dwyer 2005;Prins 2002). Minority languages, including Tibetan languages, have borrowed new words from Chinese in order to engage in new technologies and domains of use. Such linguistic processes have made Chone Tibetans feel that Tibetan languages, and in particular their own form of Chone Tibetan, are not as useful, modern or advanced as Chinese. Tashi, a grandfather in his late sixties, said: 'I always think that Tibetan is shallow. For example, in Chinese we can say the government constructs this or that thing, that it makes an airplane or cannon. Can we say these things in Tibetan?' When compared with Chinese, then, Tibetan languages -especially Chone Tibetan -do not have an indigenous vocabulary for post-1949 innovations. This lack of a modern lexicon has worked to make local Tibetans undervalue their heritage language and imagine Chone Tibetan to be 'shallow,' accelerating the language shift from Chone Tibetan to Chinese.
Language shift and the sense of linguistic inferiority are further reinforced by the linguistic market. In China, Chinese is the lingua franca and is used in almost every domain with the support of both the state and the market. Chinese is also a language that people believe can provide them with economic security and upward mobility (Bulag 2003;de Varennes 2012;Harrell 1993;Prins 2002). Tibetan is not used in the same kinds of social interactions as Chinese is, leading Chone Tibetans to perceive it as having less value. Discussing which language was more useful, Dorje Men, the former Tibetan language teacher, said, 'Tibetan is of little use. If you go to Lanzhou (the provincial capital of Ganzu province), you have to speak Chinese, and Tibetan becomes useless. Without knowing Chinese, you cannot even find a bathroom or ask people where a bathroom is.' From Dorje Men's perspective, it is almost impossible to engage in any activities in today's China without resorting to the use of Chinese. At the same time, Chinese is also valued more highly in the labor market. For most Chone Tibetans, the labor market has become a major source of income due to their gradual abandonment of livestock farming since the 2000s. Currently, only two households in Yarru village still herd sheep. The remaining Chone Tibetans either grow cash crops and medicinal herbs or choose to work as migrant workers. However, as the forty-seven-year-old Pema, who took standard Tibetan classes in the Town Center Middle school, argued, the precondition of being a migrant worker in the labor force is that 'you have to be literate in Chinese.' Increasingly, Chone Tibetans are realizing that gaining proficiency in Chinese is a prerequisite for them to lead a better -or at least modern -life and have the chance of social mobility in the context of the nation that the PRC is building. In a tangible indication of the shift, in 2002, parents in a village near Yarru even demanded that the Chone Education Bureau cancel their children's weekly Tibetan class in order that their children learn Chinese intensively and have a better chance of being admitted to a good university.
Such parental requests reflect the success of a powerful linguistic hegemony among Chone Tibetans. As Suarez has noted, 'the results of successful linguistic hegemony are often language shift from the minority language to the majority language and, ultimately, language loss ' (2002: 514). A strongly held language ideology, as the consent component of linguistic hegemony, can result in speakers of a minority language holding a negative view towards their heritage language and giving up opportunities to learn the heritage language. For example, an eighteen-year-old high school student, Khacho Men, stated firmly that Tibetan languages, especially Chone Tibetan, should not be used in formal contexts, such as meetings and conferences, because 'Tibetan is not a presentable language.' This growing sense of inferiority has bred a sense of linguistic fatalism in which local Tibetans believe that Tibetan languages will eventually cease to be used. Tashi, the grandfather in his late sixties, stated: 'not many years from now, Tibetan languages will be extinct. The influence of Chinese is so strong that it will leave no space for the development of Tibetan. Other than the monks learning Tibetan to study Buddhist scriptures, there is no use for farmers or herdsmen to learn Tibetan. [Famers and herdsmen] are incapable of strengthening the Tibetan language.' For Chone Tibetans, then, whereas Chinese is considered to be a useful language, Tibetan is perceived to be of little use on account of its low social status and slow response to new domains of use.

Establishing the Superiority of Amdo Tibetan over Chone Tibetan
Alongside perceiving Chinese to be the most useful language, Chone Tibetans also believe that standard Tibetan can necessarily only be Amdo Tibetan on account of a successful standardization movement and the use of Amdo Tibetan in formal contexts. Chone Tibetan is considered less prestigious, less useful, and less authentic than Amdo Tibetan due to its heavy reliance on loanwords from Chinese. The notion that the Chone Tibetan language has been more contaminated by Chinese at the lexical level is deeply rooted in Chone Tibetans' understandings, further contributing to Chone Tibetans' disparagement of their heritage language.
Even though in some contexts, linguistic purism can be seen as minority language speakers' resistance to a dominant language (Brunstad 2003, Turin 2014, linguistic purism can also jeopardize the maintenance of local linguistic varieties. This is because in the construction of an imagined linguistic purity, there is always a conception of what a good or standard language is. This in turn encourages people to accept and value the standard language given its support of institutional power. As a result, linguistic purism affects speakers' language ideologies and contributes to the linguistic and cultural hegemony of a standard language (Brunstad 2003). Compared to Amdo Tibetan -which is represented by institutional forces as the pure and standard Tibetan language -Chone Tibetan, with its heavy reliance on loan words from Chinese, has been belittled by its speakers as an impure and inauthentic local patois. As Woser, a man in his mid fifties, argued, 'the authentic Tibetan language should be the language spoken by Tibetans in Luchu, Machu, Labrang. 6 Their language is different from us in that we keep lots of Chinese words in our language, such as vinegar, pepper, matchstick, etc.' Loanwords from Chinese have made Chone Tibetans feel that their language is contaminated and therefore inauthentic, and they have further internalized this ideology to believe that even phonetically, Chone Tibetan is less culturally Tibetan than Amdo Tibetan. Dorje Men, the former Tibetan language teacher said, 'Amdo Tibetan is so delicate and sophisticated, while our language (Chone Tibetan) is a little bit coarse and crude.' Chone Tibetan's heavy reliance on loanwords from Chinese, plus its low social prestige due to the lack of state support, has made its speakers, such as Pema Tso, look down on their heritage language and believe that: 'Chone Tibetan is just a local dialect, while Amdo Tibetan is a [more] standardized language.' Two teenagers who had equal proficiency in Chinese and Chone Tibetan also expressed their desire to learn Amdo Tibetan, indicating that the linguistic hierarchy has taken hold in the minds of young Chone Tibetans. This hierarchy reflects not only the limited material, educational and symbolic resources 6 Luchu, Machu, and Labrang (Sangchu) are three counties of Kanlho TAP that form the nomadic center of Kanlho TAP. Since Amdo Tibetan is based on the speech of nomads, Chone Tibetans attributed a high value to the Tibetan language spoken by Tibetans from the above-mentioned three counties and view it as the authentic Tibetan language.
to which the Chone Tibetan language and its speakers have access, but also indicates how local villagers unconsciously and voluntarily participate in and reinforce the reproduction of a linguistic hegemony by putting a favored regional language (Amdo Tibetan) ahead of their own heritage language. When asked about the future of the Chone Tibetan language -from cadres to villagers, and from seniors in their eighties to teenagers -few people have an optimistic view. Woser, who received all of his education in Chinese, believes that the sidelining of Tibetan is unavoidable: 'the nature of Chinese society is Han Chauvinism. If you do not voluntarily assimilate yourself to Chinese society, you cannot go further.' Pema Tso said, 'preservation and promotion of Chone Tibetan is impossible. In the future, Chone Tibetan will only be spoken in the deep and remote valleys.' However, the majority of Chone Tibetans, including Pema Tso, are confident in the resilience and value of Amdo Tibetan. They believe that Amdo Tibetan, with the support of the state and its large number of speakers, will continue to be promoted widely. When asked if Tibetan classes would be offered in the future at the request of parents for the sake of preserving their ethnic and cultural identity in Yarru village, a government official replied, 'virtually impossible, since Chone Tibetans' linguistic environment has already become unfavorable. Nothing can be changed and Chone Tibetans are greatly Sinicized.' What remains appears to be an almost hopeless situation for the historic language of Chone. Last year, with the equivalent of USD $8,000 of funding from the office of Cultural Affairs of Gansu Province, Chone Tibetans in Yarru village started to preserve their linguistic and cultural identity through the learning and preservation of Shépa, an important local oral tradition. However, because it was a voluntary activity, only those who understood Chone Tibetan participated in the cultural activities. Also, since almost all of the participants were illiterate in written Tibetan, they learned and documented the songs by writing the Tibetan lyrics in Chinese characters based only and incorrectly on their phonetic resemblance to Tibetan words. Therefore, to what extent and how effectively Chone Tibetans can revitalize their heritage language through the learning, preservation, and revitalization of Shépa remains an open question.

Conclusion
To conclude, the language policy of the People's Republic of China has shaped the Chone Tibetan language and ideologies around its use and future. Through examining the state's language policies as they have played out in Chone post-1953 and delineating Chone Tibetans' language practices and language ideologies, it is evident that the state's coercive language policies and the understandable willingness shown by Chone Tibetans to adopt new languages work together to contribute to a threetiered linguistic hierarchy. This linguistic hierarchy positions Chinese and Amdo Tibetan above Chone Tibetan -albeit with Chinese firmly situated above Amdo Tibetan -with the result that Chone Tibetan faces a dual linguistic hegemony. This hierarchy has also shaped the language practices and beliefs of Chone Tibetans, and has led them to internalize a sense of linguistic fatalism towards their heritage language and reconcile themselves to its inevitable demise.
We hope that our research contributes to an emerging and necessary conversation about linguistic diversity within cultural Tibet, and helps to improve the visibility and viability of the varieties and dialects of Tibetan, each of which are distinct markers of identity and cultural expression. We also call for an approach to dialect documentation and preservation that is situated within the broader goal of minority language preservation in cultural Tibet. In the meantime, we hope that our work contributes methodological approaches and analytical advancements to research on minority language standardization processes, through our examination of the inherent tensions within such processes and through our discussion of how communities interact with minority language standardization projects.