Version classiqueVersion mobile

The Politics of Language Contact in the Himalaya

Selma K. Sonntag
Mark Turin

1. Language Contact and the Politics of Recognition amongst Tibetans in the People’s Republic of China

The rTa’u-Speaking ‘Horpa’ of Khams

Sonam Lhundrop, Hiroyuki Suzuki et Gerald Roche

Texte intégral

Vertical and Horizontal Politics of Language Contact in Tibet

Language contact has both horizontal and vertical dimensions. The horizontal dimension refers to the exchange of linguistic features that takes place during language contact. This includes the flows of lexicon, phonemes, syntactic structures, and so on, that occur via practices of borrowing, code-switching, and the intergenerational transmission of languages acquired in adulthood. Over time, such horizontal exchanges lead to linguistic convergence, the emergence of creoles and pidgins, and the formation of language areas. The vertical dimension of language contact, meanwhile, refers to the ordering of populations into hierarchies according to their language practices, through various processes of domination and subordination (Grillo 1989). Whilst the horizontal dimension of language contact is primarily associated with convergence, the vertical dimension is associated with linguistic differentiation (Gal 2016) and language shift — the replacement of one language by another (Pauwels 2016).

1Both horizontal and vertical dimensions of language contact influence one another, and both are inherently political. Regarding the horizontal dimension, we often see, for example, the existence of purist ideologies (Thomas 1991) underlying resistance to loanwords in accordance with the position of language varieties and their speakers in a vertical hierarchy; purism typically targets terms from threatening dominant languages, but is indifferent to borrowing from subordinate languages. The features that are exchanged in horizontal contact are also coded as indexing various types of vertically arranged categories of languages (beautiful/ugly, expressive/restrictive, etc.) and people (good/bad, superior/inferior, competent/incompetent, etc.) (Alim, Rickford, and Ball 2016; Piller 2016). The various interactions between the vertical and horizontal dimensions of language contact produce the multitude of language ecologies present today — the rich diversity of how multiple language forms are differentiated and organized in social and physical space (Haugen 2001).

2In this chapter we focus on the vertical dimension of the politics of language contact, with a discussion of ‘recognition’ as a key process through which vertical sorting takes place in language contact situations. From within the vast literature on the politics of recognition, we focus on key sources in order to introduce how this concept can be used to think about the vertical dimension of language contact. Cillian McBride (2013) distinguishes two subtly distinct varieties of ‘recognition’, which we will gloss as ‘individual recognition’, and ‘collective recognition’ — we focus on the latter. Both concepts draw on Hegel’s foundational work on the intersubjective nature of identity (Cudd 2006) — the way that individual and collective identities are formed through relationships with others, rather than arising from the inherent qualities of the individual or collective self. In this view, recognition by others is a key component of the development of self-identity. For theorists of collective recognition such as Charles Taylor and Nancy Fraser, the recognition of a person’s belonging to a larger group is foundational to healthy identity formation. Injustice arises when such identities are denied (non-recognition) or are denied equal respect to mainstream identities (mis-recognition). Both non- and mis-recognition lead to various harms, including social exclusion, economic marginalization, interpersonal discrimination against members of non- and mis-recognized groups, and a distorted sense of self and self-worth. We argue that in the context of language contact and the creation of language hierarchies, language shift can also be viewed as a harm resulting from mis-recognition.

3We examine recognition and the formation of language hierarchies in the Tibetan context. Although typically viewed as linguistically homogenous, with diversity existing only between dialects of a single Tibetan language, Tibet is actually home to significant linguistic diversity (Roche 2014, 2017). The region’s language ecology is now dominated by the national language of the People’s Republic of China (PRC), Putonghua, whilst an imagined, standard Tibetan language, represented by the written language, acts as a regionally dominant, but nationally minoritized, language. Meanwhile, this standard Tibetan language is positioned in a vertical hierarchy above the region’s spoken Tibetic varieties (Tournadre 2014) and its minority (i.e., non-Tibetic) languages (Roche and Suzuki 2018). An important factor conditioning the prestige and vitality of these minority languages is their lack of official recognition by the Chinese state, which renders them invisible to formal language policy and planning initiatives (Roche and Yudru Tsomu 2018). However, in this article we do not discuss state policies and practices, but instead draw attention to another way in which Tibet’s minority languages are subordinated in a language hierarchy — their mis-recognition by the ‘mainstream’ Tibetan population.

4We examine the mis-recognition of Tibet’s minority languages through an exploration of the case of the rTa’u language. rTa’u is spoken by approximately 45,000 people in western Sichuan Province, primarily in dKar mdzes (Ganzi) Tibetan Autonomous Prefecture. All its speakers are classified as Tibetans within the state’s ethnic classification system, and also consider themselves as such. rTa’u has a long history of contact with Tibetan, as evidenced by the numerous Tibetan loanwords it contains (Wang 1970–71). However, this contact has intensified in the past three decades in the context of rapid, state-led development and increasing human mobility. A recent investigation into the vitality of rTa’u found it to be ‘clearly endangered’ (Tunzhi 2017), with widespread language shift towards Tibetan underway. rTa’u is widely known in the linguistic literature as Horpa, and Tunzhi (2017) has argued that the use of this exonym by linguists potentially contributes to rTa’u speakers’ negative attitudes towards their language and thus also contributes to undermining the language’s vitality. In this chapter, we extend this argument in two parts. First, we examine how debates about the origin of rTa’u speakers and the ‘notorious ambivalence’ (Wang 1970–71) of the term Hor both contribute to mis-recognition, insofar as they bring into question the rTa’u speaker’s deeply-felt Tibetan identity. Secondly, we look at how this mis-recognition articulates with the broader position of rTa’u speakers in the context of contemporary debates and social movements amongst Tibetans in the PRC. We argue that this mis-recognition in two social domains — the academic and the everyday — contributes to the overall subordination of the rTa’u language within the Tibetan hierarchy of languages, which in turn is driving language shift.

5In the conclusion, we examine the implications of these arguments for understanding language contact in Tibet and the Himalaya more broadly, addressing our conclusions to both analytical and normative concerns. Analytically, we argue that the concept of recognition, despite having been critiqued by both anthropologists and linguists, is nonetheless useful in understanding the politics of language contact. Secondly, our normative conclusions examine how a more nuanced consideration of the dynamics of recognition can be used to formulate ameliorative projects that could help foster linguistic diversity in the region, and reverse the widespread language shift currently underway throughout Tibet and the Himalaya; we contrast this perspective with current local approaches to language politics in Tibet, which are, as we show below, based in essentializing and purist discourses that are likely to be contributing to, rather than resisting, language loss.

The rTa’u-speaking ‘Horpa’: Ambiguous Origins and Shifting Polysemy

A great deal of scholarly attention in the PRC has been focused on placing rTa’u speakers within the broader Tibetan community. Within these debates, the fact that rTa’u speakers are Tibetans, but speak a non-Tibetic language, is viewed as a ‘problem’ that requires solving, often through historical investigations seeking a single baptismal origin for the population, which would supposedly resolve the issue of their contemporary identity. This search for a single origin is complicated by the polysemy of the term Hor that is applied to rTa’u speakers in literary Tibetan, and which is also used in Chinese and English texts to refer to their language. We argue that the search for origins, and the polysemy of Hor, are part of a broad regime of mis-recognition that does not take into account rTa’u speakers’ professed identity as unproblematically Tibetan, despite their linguistic distinctiveness. Following, we briefly examine the main competing theories of the origins of the Horpa people before discussing the polysemy of the term Hor.

Fig. 1.1. Map of Tibet with cultural regions and prefectures. Created by the authors, CC BY.

Fig. 1.1. Map of Tibet with cultural regions and prefectures. Created by the authors, CC BY.

This map shows major locations mentioned in this chapter. The inset (left) shows the location of the Tibetan Plateau within the People’s Republic of China. The central map shows the Tibetan Plateau divided into three main cultural regions (Amdo, Khams, and U-Tsang), and also shows prefectures mentioned herein: 1) Nag chu; 2) dKar mdzes, and; 3) bDe chen. At right, dKar mdzes Prefecture is shown, with counties mentioned in the text: A) sDe dge; B) ’Ba’ thang; C) Li thang; D) Brag ’go; E) rTa’u, and; F) Rong brag.

6Numerous scholars trace the origin of rTa’u speakers to the Mongols, typically Eastern or Khalkha Mongols (Zeng 2006; Ren 1981; Gele 1988; Ganzi Xianzhi 1999; Daofu Xianzhi 1997). The Mongol invasion of what is now dKar mdzes Tibetan Autonomous Prefecture is an unequivocal historical fact. Proponents of the theory that rTa’u-speakers originate with Mongols trace the term Horpa directly to the legacy of having been invaded, and then ruled over, by Mongols, known as Hor, from the Yuan Dynasty (1271–1368) to the early Qing Dynasty (1644–1912) (Tunzhi 2017). Although no linguist today considers rTa’u to be a Mongolic language, there has nonetheless been speculation that rTa’u is related to Mongolian, such as in Zeng (2007: 186):

We departed from Dajianlu (Kangding), setting out to explore the Hor region. As we passed through Songlinkou, we discovered that the Horpa people are different from Khams-Tibetans. Our interpreter, who was from Kangding told us […] that the language spoken by Horpa people is a mixture of Mongolian and Tibetan called Dijiaohua, which is incomprehensible to neighboring Tibetans.

  • 1 For instance, the township located about ten kilometers east of Luhuo County town called S (...)

Zeng (2007) further cites reference to local toponyms in identifying Mongolian influence.1

  • 2 Ran (2004: 26–37) dates the antiquity of this settlement to approximately the fifth and si (...)

7A second theory that seeks to explain rTa’u origins is what we call the nativist theory. This theory has received little attention — only a few scholars from dKar mdzes Prefecture have written on the topic, though it is widely discussed in local intellectual circles. A key proponent of this view is Ran (2004), who, in addition to his published work, has also given several oral presentations on the theory, and has written a number of unpublished manuscripts on the topic. The nativist theory argues that place names in the Tibetan-speaking area to the west of the rTa’u-speaking region provide good evidence that the rTa’u language predates the arrival of the Tibetan language. Ran (2004: 27) provides a list of mountain and place names which he argues are rTa’u in origin, e.g., that use the prefix /ʐæ/, meaning mountain in rTa’u, but having no meaning in Tibetan. This is significant given that rTa’u language is today no longer spoken in this area. Ran (2004: 26–37) thus argues that rTa’u speakers or the rTa’u language are native to the area,2 and prominent local Tibetan scholar Thubtan Phuntshog (Thub bstan Phun tshogs, p. c.) concurs with Ran’s argument.

8Although situating rTa’u speakers as indigenous might be viewed as an attempt to raise their status within local language hierarchies, like the theory of Mongol origins, it still constitutes a form of mis-recognition in treating Tibet’s minority language speakers as a problem that needs to be solved — an aberrance from an assumed mainstream population — and in not taking seriously speakers’ professed identities. Such debates about the origin of rTa’u speakers can be understood in the broader situation of the numerous ways in which the term Hor is used in Tibetan contexts. An exploration of how the meaning of this term has changed over time, and shifts according to context, will show that the mis-recognition found in the search for Horpa origins is also perpetuated by the use of this polysemous label.

9The word Hor appears in both written and spoken varieties of Tibetan. For instance, Powers and Templeman (2012: 299) define Horpa as: ‘a term used by Tibetans to refer to Uyghurs, and sometimes to Mongols. It generally refers to Turkic or Mongolian people living near Tibetan populations in northern Tibet and modern Qinghai.’ Edgar (1932: 71) states in a description regarding the people called Hor in the dKar mdzes area that, ‘the “Yugurs” and the “Hor” are the same people.’ In the following discussion of the term Hor, we first examine and compare definitions in several modern Tibetan dictionaries, and then, with reference to Moriyasu (1977), explore the shifting meaning of Hor over time. Finally, we survey the different meanings of Hor in a variety of local contexts throughout the eastern Tibetosphere.

  • 3 Li thang in Literary Tibetan. However, local Tibetans in Li thang consider the toponym as (...)
  • 4 Hor sras khag lnga ‘five districts ruled by the Hor prince’ in Literary Tibetan.
  • 5 Hor dpon khag lnga ‘five Hor chieftain’s states’ in Literary Tibetan. There are also other (...)
  • 6 Each of them in Literary Tibetan is rTa’u, Brag ’go, Tre hor, dKar mdzes, and sTong skor. (...)

10Amongst modern Tibetan dictionaries, Jäschke (1881: 598) defines Hor according to two historical periods: formerly, it designated a Mongol, whilst at the time of writing it referred to Tibetans living near Lake Nam (gNam mtsho) in Central Tibet, or to Turkic peoples in Western Tibet. Les Missionnaires Catholiques du Thibet (1899: 1066) first defines Hor as a term for both regions and people, namely, in the western Tibetan area it refers to Turkestani people, whereas in central Tibet it denotes various pastoral tribes in the North Plains (Byang thang) region. Additionally, Hor is also claimed to designate Le thang [sic],3 ’Ba’ thang, and sDe dge, as well as a large region referred to as the Hor zar Khag lnga [sic]4 or Hor po Khag lnga [sic].5 Das (1902: 1329, 1330) defines Hor as ‘a Tatar’, and Hor pa as ‘a Dzungarian; also a Tibetan from the northern provinces, a herdsman of North Tibet.’ Zhang (1985: 3071–72), meanwhile, defines Hor as describing various ethnic groups in different periods, i.e., Yugur (Uyghur) before the Yuan; Mongol in the Yuan; ’A zha and Tuyuhun in the Yuan-Ming transition; and at present, pastoralists in the Byang thang, and members of the Tu nationality (minzu) of Qinghai. Zhang also provides two additional meanings: a po hor, for pastoralists living in northern Tibet, and people of the Khams Tre Hor region, although the Chinese description in the dictionary is ‘Mongolians living in five regions: Daofu, Luhuo, Zhuwo, Ganzi, and Donggu’.6 Goldstein (2001: 1175) provides two meanings: firstly, referring to either Mongolia or Mongols or, secondly, to the pastoralists living in Northern Tibet.

  • 7 Not giving a detailed description on the usage of hor in the western Tibetosphere within t (...)

11We see both variation and overlap in these definitions. Whereas Goldstein (2001: 1175) does not distinguish Mongols and pastoralists in Northern Tibet, Jäschke (1881: 598) clearly mentions the temporal distinction between these two; however, as a modern synchronic description, Hor in the sense of Mongol also appears in compounds such as Hor glu ‘Mongol song’, Hor gos ‘Mongol clothing’, and Hor zla ‘Mongol calender’. Jäschke (1881: 598), Les Missionnaires Catholiques du Thibet (1899: 1066), and Zhang (1985: 3072) point out that the meaning of Hor differs according to region, denoting pastoralists of northern U-Tsang, Turkic people in Western Tibet,7 and people in five Hor districts in Khams. Therefore, Hor is a polysemic word when viewed synchronically. Finally, as Goldstein (2001: 1175) makes clear, the meaning also varies depending on whether a text is in Classical or Modern Tibetan.

  • 8 This collection of documents, found in the Mogao caves in contemporary Dunhuang, Gansu Pro (...)
  • 9 The transliteration of this document is available at (...)

12Moriyasu (1977) provides a diachronic perspective on the polysemy of the term Hor through a detailed analysis of the Dunhuang document,8Pelliot Tibétain 1283,9 together with other primary sources in Chinese, Old Uyghur, and Literary Tibetan. He argues that Hor originally denoted non-Tibetan groups living on the northern Tibetan plateau, regardless of their specific ethnicity, but excluding the Han of the Tang and Qing Dynasties, and the Uyghur (1977: 43–44). He divides changes in the object denoted by Hor into five periods (see Table 1.1):

Table 1.1. Changes in the meaning of Hor over time.

Era Meaning of Hor
7th c. to mid 8th c.
(until early 9th c.)
ethnic groups in northern Tibet, between Tibet and Tang
late 8th c. to 9th c. equivalent to Dru gu (Turkic or Uyghur)
13th c. to 15th c. Mongol
(16th to) 17th c.
(until 18th c.)
ethnic groups of Qinghai and Turkic and Mughal people in Western Tibet
19th c. to 20th c. ethnic groups of Northern Tibet (Byang thang) and Eastern Tibet (sDe dge)

Source: Moriyasu 1977

He concludes by defining Hor as ‘ethnic groups except for the Han living on the northern Tibetan plateau, as well as those living in contact with the border of Tibet’ (Moriyasu 1977: 45). Moriyasu thus claims that for Tibetans, Hor denotes non-Tibetan ethnic groups on the northern Tibetan plateau, which is why it excludes Han, who are the counterpart to the east of Tibet. Based on this conclusion, we can understand that the polysemy of the word Hor is partially due to an accumulation of various meanings, and the meaning denoted by the word depends on the era and region.

  • 10 In the earlier period, sog denoted countries and states to the west of Tibet, mainly the p (...)
  • 11 rGya bod hor sog gyi lo rgyus nyung brjod pa byis pa ’jug pa’i ’bab stegs.
  • 12 dBal mang paN Di ta dKon mchog rGyal mtshan (1764–1863). Although Oidtmann (2014) transcri (...)

13If we turn to contemporary usages of the term, we find significant differences from this historical situation. Moreover, there are differences between Hor as an autonym and an exonym. We will examine how Hor is used in the regions of U-Tsang, Amdo and Khams, following an introduction to Hodgson’s (1853) distinction between Hor and Sog.10 Hodgson (1853: 122–23) describes Hórsók as a compound of two Literary Tibetan words hor and sog; however, no modern Tibetan dictionary we consulted includes this compound (Jäschke 1881; Les Missionnaires Catholiques du Thibet 1899; Zhang ed. 1985; Goldstein 2001). Oidtmann (2014) also discusses Hor Sog as a copular expression appearing in the title of a text11 by Welmang Pandita Konchok Gyaltsen,12 and analyses Hor Sog as ‘Hor and Sog’, referring to the Yuan and then later Mongol khanates, especially that of Gushri Khan (2014: 306). Both Ahmad (1970: 110) and Tucci (1999: 256n128) make a similar distinction between Hor and Sog as representing distinct Mongol populations.

  • 13 Zhang (1985: 3072, 3121) provides a spelling a po hor, which might follow the pronunciatio (...)
  • 14 This group was previously named Kham-Hor in Tournadre (2005).

14In the context of the U-Tsang, Hor primarily denotes Tibetan people living in the Nag chu region, and is used as both an exonym and autonym; people in Nag chu call themselves Hor pa, whilst outsiders call them a pho hor,13 an appellation that typically carries a derogatory connotation. Nag chu Tibetans nonetheless express strong attachment to their identity as Hor, and in particular to their place within the thirty-nine Hor tribes, Hor tsho So dgu (Karmay 2005; Shi and lHa mo thar 2012). Linguistically, the Nag chu Hor language is considered a member of the Khams pastoralists’ dialect group by previous works such as Qu (1996) and sKal bzang ’Gyur med and sKal bzang dByangs can (2002), and as a member of the southeastern section together with many Khams Tibetan subgroups (Tournadre 2014).14

15In Amdo, the northeastern part of the Tibetosphere, Roche (2011) mentions that Turkic-, Sinitic-, and Mongolic-speaking peoples of Amdo are often construed as and referred to as Hor, and he documents how Tibetans in Xing’er Township (Minhe County) refer to local Mangghuer people as Hor. Nowadays, many Amdo Tibetan speakers primarily understand Hor as a term designating the people of Tu nationality (hor rigs). However, in referring to the Tu as Hor, the term Hor pa is not used in Amdo.

16In Khams, especially within the present dKar mdzes Prefecture of Sichuan Province, we find Hor as an appellation of ethnic groups as well as toponyms. According to Moriyasu (1977), Tibetans in Central Tibet began, in the nineteenth century, to use Hor to refer to inhabitants of the ‘five Hor regions’: rTa’u, Brag ’go, Tre hor, dKar mdzes, and sTong skor in today’s dKar mdzes (cf. Zhang 1985: 3071); inhabitants of this region are referred to as Hor pa, and toponyms containing Hor are widespread. However, it is crucial to note that Hor in this context is an exonym. Although some contemporary Tibetans of the ‘five Hor regions’ now refer to themselves as Horpa, the term appears to have originated as an exonym approximately 200–300 years ago, in the Yongzheng period of the Qing Dynasty, i.e., the early eighteenth century (Li 2015: 120–21).

  • 15 Huorong. Here and following, the Chinese pinyin for Tibetan place names are provide in the (...)
  • 16 Honglong.
  • 17 Huoqu.
  • 18 Heni or Heranniba.

17The extent of the Hor khog Khag lnga is limited to the northern part of the contemporary dKar mdzes Prefecture, from dKar mdzes County in the north to rTa’u County in the south. However, we can find toponyms containing Hor even in southern Khams, up to the present gTor ma rong valley in bDe chen Prefecture, where we find a hamlet called Hor gzung,15 which locals interpret to mean ‘the place grasped (controlled) by Hor’. Tibetans living in this hamlet consider Hor to mean ‘Mongol’, referring to Mongols who came to the area during the Yuan Dynasty. Li thang and Nyag chu kha counties also have several toponyms such as Hor lung (township),16 Hor chu (river),17 and Hor rnying a.k.a. Hor ra rnying ba (township).18 Locals also interpret these names to be connected to Mongols of the Yuan Dynasty.

18Throughout Khams, the term Hor (for Mongols) is typically conflated with Sog. In interviews with local Tibetans residing in or near hamlets containing Hor, the term Sog was consistently used to refer to Mongols, not Hor. No one was able to explicitly distinguish Hor from Sog. Some outsiders suggest that Hor in Li thang and Nyag chu kha can be interpreted as Hor in Hor khog Khag lnga; however, except for the mention of Les Missionnaires Catholiques du Thibet (1899: 1066) that Hor also designates Le thang [sic], ’Ba’ thang, and sDe dge, nowhere else is it recorded that Li thang and Nyag chu kha were part of Hor khog Khag lnga. Furthermore, we can note that in some toponyms in dKar mdzes Prefecture, Sog often refers to ‘Mongol’, even in relation to the Yuan Dynasty, such as Sog pho (township) of Rong mi Brag ’go (a.k.a. Rong brag) County. Local Tibetans trace their origins to Mongol soldiers who did not complete the journey to Yunnan. Contemporary residents are Khams Tibetan speakers, and are generally unfamiliar with the term Hor. Lastly, we refer to Giraudeau and Goré’s (1956) French-Tibetan dictionary, which primarily reflects spoken varieties from the southern Khams area. They give a single Tibetan word for ‘Mongolian’: sog po, followed by an example Sog po gong ma, denoting ‘Mongolian or Manchurian emperor’ (1956: 185). This means that they understand Sog po both as Mongolian and Manchurian in the historical context, whereas it only means Mongolian in the contemporary context. This view, except for the sense of Manchurian, is widely shared by Tibetans in southern Khams today. The word Hor as a designation of Mongol does not appear in this dictionary.

19To summarize, then, local Tibetans in Khams often understand both Hor and Sog to mean Mongol, but use Sog for Mongols in the modern age, and Hor in a historical context related to the Yuan Dynasty. As Moriyasu (1977) points out, Mongolians in the Yuan Dynasty were called Hor by Tibetans; however, Tibetans in the contemporary era hardly recognize the historical difference between the two lexemes, and thus understand them as synonyms.

  • 19 The following section is based on two periods of fieldwork, in 2014 and 2017. Data were co (...)

20Taken together, the debates around the origins of rTa’u speakers, and the polysemy of the term Hor suggest the same thing — that rTa’u speakers are somehow less than Tibetan; that their identity is a problem that needs to be solved, since they are distinct from a perceived Tibetan norm. Both the search for origins and the exonym Hor suggest that somewhere there is an unproblematic, standard Tibetan to whom rTa’u speakers are being compared. The polysemy of the term Hor is particularly telling in this regard, since the only point at which all possible meanings of the term converge is around the concept of a ‘non-Tibetan’ inhabitant of Tibet. Meanwhile, the debates around the origin of rTa’u speakers enact mis-recognition in a more subtle way, in that, even when finding rTa’u speakers to be indigenous, they serve the ‘metacultural’ function (Urban 2001) of validating a link between historical origins and contemporary identity, and thus overriding the views and beliefs of contemporary rTa’u speakers, whilst also perpetuating the status of rTa’u speakers as a problematic population that necessitates special explanation for their distinctiveness. These academic infatuations with the supposedly problematic nature of rTa’u speakers and the Hor stand in stark contrast to rTa’u speakers’ self-perception as unproblematically Tibetan, despite their linguistic distinctiveness. Even though most rTa’u speakers are unaware of these debates — only two college graduate interviewees out of a total of thirty had even heard the term Hor19 — we nonetheless argue that these debates are significant in that they form part of a larger regime of mis-recognition when viewed in the broader context of contemporary discussions around Tibetan identity and language in the PRC.

rTa’u-speakers and Contemporary Tibetan Language Politics

Within contemporary discussion around language and identity amongst Tibetans, minority languages such as rTa’u are faced with a pervasive climate of mis-recognition. This primarily manifests as the assertion of monoglot nationalism (Dorian 1998; Heinrich 2012), a standard language ideology (Lippi-Green 1997) that portrays all Tibetans as speaking a single language. This singularity is often explained in terms of common origins and descent, with all the languages of Tibet typically viewed as rooted in the written form. In this view, non-standard languages are thought to be corrupted, degraded forms of the written language, as expressed in the following quote from the Tibetan historian Tsepon WD Shakabpa (2009: 13).

The spoken and written forms of Tibetan are closely related, although there are modest corruptions in the spoken form in widespread areas: Lhodruk, Sikkim, Ladakh, Monpa, Sherpa, Tamang, and so forth […] although there is one original language for the three provinces of Tibet, regional accents have evolved. Because of these corruptions, if one does not listen carefully, it is difficult to understand.

  • 20 Teichman then goes on to discuss the ‘curious racial mixture’ of local Tibetans, suggestin (...)

Such a view portrays rTa’u and other minority languages of Tibet as corrupted forms of written Tibetan, thus placing them beneath the written language in a prestige hierarchy organized around the principle of purity. This view of non-standard languages as degraded forms of the written language can be gleaned from Teichman’s (1922: 65, emphasis added) description of rTa’u, which most likely reproduces the views of local interlocutors: ‘The language spoken at Dawu (perhaps akin to that used in the Gyarong States further east) is a very corrupt form of Tibetan, if indeed it is a Tibetan dialect at all.’20 Similarly, Rockhill’s (2004 [1891]: 263) description of rTa’u language as ‘a wonderful mixture of Tibetan and Chinese’ is most likely derived from his local informants, and represents the Tibetan standard language ideology that diversion from an ancestral language comes about via ‘corruption’ from other languages.

  • 21 A speech on Mdo Kham rTa’u area by ‘gyaur Med Tshering, 26 April 2014: http://ti.zangdiyg. (...)
  • 22 A short take on the relationship between rTa’u language and written Tibetan by rTa’u rGyal (...)

21When not tied to the written language by a process of temporal degradation, minority languages such as rTa’u may be connected to written Tibetan by an imagined process of fossilization; divergence as stasis rather than change. A number of online commentaries21 have attempted to establish some sort of relationship between Old Tibetan and contemporary rTa’u, in support of the theory that rTa’u is a ‘living fossil’ of Old Tibetan. One online article22 by a rTa’u-speaker suggests that rTa’u speakers are the descendants of soldiers sent by Tibetan kings to safeguard the eastern entrance to the Tibetan plateau. According to the author, this explains the supposed character traits of contemporary rTa’u speakers: negotiation skills and a tendency to fight. The author even suggests that the language itself was developed as a secret code for spying purposes. This article, from a native perspective, if nothing else, emphasizes the desire of rTa’u people to be identified as Tibetans.

  • 23 The Tibetan & Himalayan Library, Tibetan text archive, English definition of logs skad, ht (...)
  • 24 Suzuki and Sonam Wangmo (2016) discuss the term logs skad.
  • 25 In addition to ‘ghost’, other suggested translations for ‘dre include ‘demon’, ‘imp’, ‘gob (...)

22Alongside such attempts to identify rTa’u with Old Tibetan, local discourses also draw attention to the language’s radical difference from spoken Tibetan. When referring to the language, non-rTa’u speakers typically refer to it not as Horpa (as in Literary Tibetan) or rTa’u (as speakers refer to it), but as logs skad — a widespread Tibetan term used to refer to non-standard languages in the eastern Tibetosphere. This term has been defined as referring to a ‘kind of speech not understood by others in a certain region, line of work, etc.’23 And although this definition appears merely descriptive and non-pejorative, rTa’u speakers consider logs skad to have definite negative connotations; it is not a term they themselves use.24 Another term applied by other Tibetans to the rTa’u language, but not used by rTa’u speakers themselves, is ’dre skad, which literally means ‘ghost speech/language.’25 As with logs skad, this term carries negative connotations. Furthermore, these derogatory terms for the rTa’u language are generally associated with broadly negative stereotypes regarding its speakers: that rTa’u speakers are shrewd business people (and are therefore by implication greedy and dishonest), have a tendency to fight, and have no appreciation for (mainstream) Tibetan culture.

  • 26 For further discussion of this system, see Chapter 5 of this volume, by Bendi Tso and Mark (...)

23These portrayals of Tibetan as a prestige ur-language, and rTa’u as either a degraded or fossilized (but always radically different) derivative, take on heightened significance in the contemporary context of widespread social mobilization in defense of the Tibetan language. Although the vitality of Tibetan is certainly greater in comparison to the region’s minority languages, concern for, and mobilization in defense of, the Tibetan language has intensified in the twenty-first century (Robin 2014; Roche 2017). Such concerns have emerged, in particular, in responses to changes in schooling, which have increasingly seen the promotion of what is called the ‘type two educational model, ’26 which has basically instituted Chinese-medium education and relegated Tibetan to a subject, rather than a medium of education (Henry 2016). In response to this, and other pressures on the Tibetan language, two major programs have emerged as part of a broader language movement: one promoting the use of ‘pure father-tongue’, (pha skad gtsang ma) and the other advocating grassroots literacy.

24The ‘pure father-tongue’ movement is a form of ‘verbal hygiene’ (Cameron 1996) aimed primarily at avoiding loanwords in spoken and written Tibetan, and promoting Tibetan neologisms (Thurston 2015, 2018). The underlying logic of the movement is that the Tibetan language is threatened by an increasing number of loanwords from Chinese, particularly as the language expands into new domains. Rather than a discrete, organized social movement with a clear organizational structure and program of activities, the pure-father-tongue movement is a diffuse, decentralized, grassroots movement (Roche and Lugyal Bum 2018). It is promoted through social media and in essays, poems, memes, and songs, as well as by word of mouth. The following examination of pure-father-tongue discourses is suggestive of the ways in which they constitute a form of mis-recognition for speakers of rTa’u and other minority languages, in positing an essentializing link between language and identity that elides the existence of Tibetan minority languages and speakers.

  • 27 See the poem online here, (...)
  • 28 Here and elsewhere, the Tibetan texts have been transcribed faithfully according to the or (...)

25One can easily find many poems dedicated to the pure father-tongue, as in the following sample by Tibetan poet Pedma Trashi (Pad ma bkra shis).27 In this poem, the Tibetan language is considered a cornerstone of Tibetan identity, along with other important elements such as being compassionate and abstaining from alcohol. Such poems are recited at school ceremonies and other social events. A common theme in all such poetry is that language is the keeper of culture, and that without culture one group of people is no different from any other, and therefore the key to maintaining Tibetan identity is the maintenance of a pure father-tongue.28

ཨོ། ཁ་བ་ཅན།
ཨོ། བྱམས་སེམས་ཅན་གྱི་མྱི་རྱིགས་ཡྱིན།
ཨོ། ཆང་དང་ཨ་རག་ཁེད་ཀྱིས་བཏུང་བ་མྱིན།

  • 29 Translation by Sonam Lhundrop.

Oh, people of the Snowland
Oh, people of compassion
Oh, you forbid alcohol
Women’s personality is softer than wool
Men honor dignity the highest of all. Oh!
Wear authentic Tibetan clothes, good sons and daughters,
Speak pure Tibetan language, people of good lineage.29

  • 30 In addition to the two songs described below, another extremely popular song on this theme (...)
  • 31 You can view Gergyal Pedma performing the song here: (...)

The most popular way that such ideas reach the general population is undoubtedly through song. Many songs in recent years have highlighted the importance of speaking pure Tibetan.30 In 2015, a song titled ‘Father-tongue (Pha skad)’ was the main theme at the New Year Gala of Khampa TV, which featured the young Tibetan singer Gergyal Pedma (dGe rgyal Pad+ma) singing the following lyrics:31

ཕ་སྐད་། ངེད་ཚོང་གྱི་ཕ་སྐད།
ཕ་སྐད། ཕ་སྐད། ངེད་ཚོང་གྱི་ཕ་སྐད།

  • 32 Translation by Sonam Lhundrop.

Nourished in the wealth of history,
The melodious sound,
Created by the unvanishing spirits of ancestors,
Embodiment of the long-lasting tales of our ancestors,
Father-tongue — our father-tongue,
Sweeter than sweet tsampa,
Father-tongue, father-tongue — our father-tongue.32

  • 33 See the song performed here:

Another clear example of such promotion is the song ‘Manifestation of the Father-tongue (Pha skad kyi Rang mdangs)’ by the well-known singer Shertan (Sher bstan):33


In my heavenly land of snow
High above shine the sun, moon and stars
Within, blessings rain down —
The fortunate blessings of the Tibetan people


In the palace of monastic universities
Great teachings pile up like a great mountain
Within, a great ocean of wisdom swirls —
The fortunate blessings of the Tibetan people


To the benevolence of the great minister Thunmi
The Tibetan letters shine with glory
Within, the flowers of great civilization blossom —
The fortunate blessings of the Tibetan people


My father-tongue is as pure as gold
My writing system is as pure as fresh milk
Within us, unity is pure as a white silk strip —
The fortunate blessings of the Tibetan people


  • 34 Translation by Sonam Lhundrop.

Pure language is Tibetans’ treasure
The letters of the alphabet are Tibetans’ adornments
Within, the cultural manifestation reflects who we are —
Do not lose that which the sky has protected34

Some have even seen it as a marketing strategy to include such songs in albums so that they reach even larger audiences. In the music industry, a new term has been invented in light of this, phake chedruk (pha skad ched bsgrigs) ‘pure-father-tongue album’, meaning that such albums include only Tibetan language songs.

26Although informal groups sometimes form to encourage people to speak ‘pure’ Tibetan, it is more often promoted by individuals, particularly those with some level of formal education. One impact of the movement has been to establish a value hierarchy that promotes ‘pure’ language as both prestigious and morally valuable, and hence a source of pride, and denigrates ‘mixed’ language as non-prestigious, immoral, and shameful. This is significant for minority languages such as rTa’u in that, as we have seen above, they are typically viewed as mixed languages. Television has been a popular medium for propagating such purist language ideologies: an example will suffice to demonstrate. In one well-known clip, a pastoralist goes to a city to purchase goods for the upcoming Tibetan New Year. He enters a shop and starts speaking a ‘mixed’ language, assuming that the shopkeeper must be Han Chinese. To his surprise, the shopkeeper is a Tibetan. Nonetheless, instead of speaking Tibetan to the shopkeeper he continues in ‘mixed’ language, now thinking to impress her and show his superiority. To his disappointment and embarrassment, the shopkeeper reprimands him and lectures him on the importance of speaking ‘pure’ language (Thurston 2015 discusses the role of comedy in propagating language ideologies in the Tibetan context).

27In addition to this focus on verbal hygiene, the contemporary Tibetan language movement has also focused on adult literacy. People from different backgrounds, such as college students, monks, and business people, have been organizing village-level programs to teach written Tibetan to illiterate pastoralists and farmers. These programs have also appeared in the rTa’u area amongst rTa’u speakers, including in Tunzhi’s village. His mother is sixty-five years old at the time of writing, and has never been to school, but has participated in a village-level illiteracy eradication program since 2014 (each household is expected by the monastery to send at least one member to attend these classes). She is now able to read several common Buddhist scriptures and chant them while following a recording on audiocassette. Cases such as hers are now common in the rTa’u area. And although we acknowledge that such programs are very much welcomed by local communities, we argue that they contribute to the mis-recognition of rTa’u speakers insofar as they reaffirm the essentializing, monoglot link between language and identity, whilst also undermining the perceived value of the spoken language and precluding efforts to develop a writing system for it.

28Also known as the ‘eradicating illiteracy’ (yig rmongs sel) program, the adult literacy program aims to teach adults basic Tibetan literacy so that they can independently read Buddhist scriptures. Each monastery oversees a specific district, and conducts such programs in communities within its district. Often the monastery develops the textbooks. Classes are continued until participants are able to recite a designated scripture. No comprehension lessons are offered; the pedagogical content simply includes how to pronounce the letters, how to combine them into syllables, and some provide chances to practice reciting simple texts. A typical program lasts about 15–20 days and is carried out annually prior to the New Year, a time when all residents are back in the community from their seasonal work at construction sites far away. In the case of the first author’s community, the local monastery sends a head teacher who teaches the advanced classes, while local college students who are back in the community during the winter holiday teach the beginners. Students are divided into different classes based on their reading proficiency. During class, students repeat letters and portions of text after the teacher. There are simple tests at the end of each program, in which students recite a given text. However, there are reports and pictures on the social media platform, WeChat, of other communities with more advanced and long-term classes, and videos circulated on WeChat show adults memorizing long Buddhist texts and demonstrating their mastery of the content through debate. In many cases, college students and local monks collaborate on such programs.

29Pressure to learn written Tibetan and to speak Tibetan in a manner that conforms to the written standard is increased by the language’s association with Tibetan Buddhism and religious figures. Many rTa’u speakers describe feeling anxious when lamas are present, especially when they are invited to perform rituals in family homes, because such figures sometimes scold locals who speak Tibetan with an accent, or even simply for speaking rTa’u in their presence. For rTa’u-speakers, their identity and sense of belonging to a community are strongly linked with their religious affiliation, and therefore such pressures from religious figures, combined with Buddhist institutional support for purism and literacy, are felt as a heavy burden. For many rTa’u speakers, such pressures are viewed as a choice between their language, on the one hand, and their religion and identity on the other.

30In this context, literacy in Tibetan becomes a nexus for maintaining ethnic boundaries through purist practices, promoting alternative visions of modern Tibetan identity through literacy, and accessing the sacred. It is perhaps not surprising, then, that when asked if there is any value in maintaining the rTa’u language, most interviewees directly compare its value with the maintenance and development of the Tibetan language. They claim that maintaining rTa’u is valuable only insofar as it contributes to the maintenance of the Tibetan language. Otherwise, the maintenance of the language for its own sake is seen as meaningless. One particular interviewee, the head of a school for orphans in the rTa’u area (and therefore the bearer of significant responsibility for the transmission of the language) stated that one language is enough for Tibet, and all effort should be directed towards the development and promotion of a single Tibetan language, even at the expense of the rTa’u language. This is demonstrative of the hegemonic position of monoglot language ideologies in relation to contemporary visions of Tibetanness.

31This unwillingness to maintain rTa’u manifests in resistance to efforts to develop the language, for instance, by creating a writing system for it. This contributes to a vicious cycle where the language is seen as useless and therefore not worth maintaining, but cannot be made useful without such initiatives. This can be seen particularly in the educational context. Although most rTa’u speakers are unwilling to develop a writing system for their language, the lack of a writing system can be seen as one of the factors that deprives rTa’u of social and cultural prestige, and places pressure on speakers to abandon their language for Tibetan. Children are encouraged to speak Tibetan at school (even when the majority of students are rTa’u speakers), and only use rTa’u at home.

32However, even this last bastion of the language is now under pressure, as it is commonly thought that speaking rTa’u hinders one’s capacity to learn Tibetan well, as rTa’u is seen as a corrupting influence on pure Tibetan. However, the challenges faced by rTa’u speakers in learning Tibetan can more accurately be explained by the failure of the educational system to account for the fact that they are learning a second language, not ‘their own’ first language. Nonetheless, prevailing views amongst rTa’u speakers blame the children themselves for being poor students, and also blame the rTa’u language for ‘corrupting’ the Tibetan language. Such views have convinced rTa’u-speaking parents that speaking Tibetan at home with their children is in their best interests.

33One interviewee in particular shared a story that reflects the motives that lead rTa’u speakers to experience persistent linguistic anxieties in relation to their spoken Tibetan. She is from a community where all households but hers speak rTa’u, so she has become fluent in both her home language, pastoralist Tibetan (’brog skad), and rTa’u since an early age. However, she said that since she moved to the city of Chengdu a few years ago due to health issues, ‘Other Tibetans say I speak Tibetan with an accent… this made me very embarrassed.’ For the vast majority of rTa’u speakers, such moments of mis-recognition based on their accent are incidents of great embarrassment and shame; rather than being recognized and esteemed as multilingual Tibetans, rTa’u speakers are thus mis-recognized as aberrant, and disesteemed.


Above, we have discussed how mis-recognition of rTa’u speakers functions in two distinct arenas: the academic and the quotidian. Within the academic field, knowledge production focuses on attempting to clarify the origins of the ‘Horpa.’ The ‘notorious ambivalence’ (Wang 1970–71) of this term, both diachronically and synchronically, suggests that its only stable function has been to imbue various populations with contrastive alterity, as Others to the Tibetan Self. Meanwhile, the everyday lived reality of rTa’u speakers is increasingly characterized by mis-recognition within the context of an essentializing language movement that seeks to tie Tibetan identity to a single, ‘pure’ language. In both cases, rTa’u speakers are denied recognition by significant social others — Tibetan academics, clergy, and laymen. The claims of rTa’u speakers to be Tibetan whilst speaking a non-Tibetan language are at best problematized, and at worst outright rejected, resulting in various harms for rTa’u speakers, including linguistic anxiety and insecurity, ultimately leading to language endangerment and shift.

34We believe that the preceding discussion demonstrates the analytical value of applying the concept of recognition to examinations of how hierarchies emerge in language contact situations. However, we acknowledge that not everyone shares our enthusiasm for the analytical usefulness of the concept. Scholars such as Coulthard (2008, 2014), Povinelli (2002), and Vincent (2017), for example, have argued that although a focus on recognition helps reveal the existence of diffuse and often concealed symbolic value hierarchies, it nonetheless leaves power hierarchies in place. These authors claim that since recognition is usually given by dominant groups to the subordinated, it continues the disempowerment of minoritized groups and thus both perpetuates and conceals existing power structures, rather than eradicating them. Simpson (2014) has offered the concept of refusal as an alternative to the asymmetrical and disempowering notion of recognition; to refuse is to reject the power structures inherent in recognition, and to assert Indigenous identities and practices on the basis of continuing sovereignty, rather than intersubjective processes that entangle the Indigenous subject with practices of settler dominance. In reply to such criticisms, we note that these analyses were all developed in Western, settler-colonial, liberal-democratic, multicultural societies, where those offering recognition (and perpetuating mis-recognition) are viewed by subaltern populations as dominating, oppressive Others. In the present case, however, we have been examining a situation in which a linguistic minority is facing mis-recognition from a dominant, mainstream group with which it identifies, and which it does not wish to refuse. As such, appropriate recognition is likely to be deemed by them not only apt, but, given the important role that religion plays in setting the boundaries of the desired collective belonging, also morally good. This therefore opens up the possibility that structures of recognition and mis-recognition, whilst always implying power asymmetries, do not necessarily imply oppression.

35A second line of argument that should be considered here relates specifically to recognition and languages. Although typically not discussed explicitly in terms of recognition, or with reference to authors such as Taylor and Fraser, there has been ongoing debate amongst linguists for at least the past twenty-five years regarding the value of ‘differentiating’ languages as bounded, stable ‘objects.’ We regard this conversation to be, in important ways, about recognition, and not only the differentiation of linguistic types and the creation and reinforcement of symbolic hierarchies. Criticism of the politics of recognition as it relates to language include: that such practices can impose inappropriate, etic standards for what constitutes a language (Mülhäusler 2006); that it may lead to the commodification of and competition between languages (Dobrin, Austin, and Nathan 2007); and that it often reproduces standard language ideologies and essentialist theories of language and identity that underlie much language endangerment (Heller and Duchêne 2007). Such views are also bolstered by broader poststructuralist critiques that view language as a fundamentally dynamic, fluid, fuzzy set of resources (Makoni and Pennycook 2005), rather than as a preexisting structure to which individuals belong; in this view, the assertion of stable identity categories, rather than their organization into hierarchies, is the source of injustice. In response to these criticisms, we note that the existence of distinct linguistic types is not in question in the context we are examining; rTa’u speakers recognize the distinctiveness of their language from that of mainstream Tibetans, who, in turn, recognize the distinctiveness of rTa’u. To refute these salient local categories would thus be to engage in both ‘epistemic violence’ (the rejection of the subaltern subject’s capacity to know, see Spivak 2010), as well as ‘ontological violence’ (violence against the conceptual order and lifeworld constructed by subaltern subjects). In contexts where discrete, bounded languages are already part of local views, we therefore see discussions of recognition as analytically appropriate.

36Beyond the analytical value of engagement with the concept of recognition in understanding the construction of language hierarchies, and in nuancing our understanding of recognition in different contexts, we also argue that the concept has a normative role to play in efforts to support individual languages and maintain linguistic diversity. We argue that greater recognition for the rTa’u language by the mainstream Tibetan community would help in its maintenance, and we therefore make the following recommendations. Firstly, academics in the PRC and elsewhere who speculate about the identity of rTa’u speakers should take seriously the expressed opinions of rTa’u speakers. Secondly, given that the polysemy of the term Hor reduces its discursive role to something like ‘un-Tibetan’, scholars would be well advised to avoid this term. Thirdly, proponents of adult literacy and the ‘pure-father-tongue’ movements could give greater recognition and esteem to minority languages such as rTa’u in relation to the ‘standard’ Tibetan that they promote. Finally, the Tibetan community at large, including local religious figures, could contribute to this increased recognition for the language by avoiding negative terms such as logs skad and ’dre skad, and by valorizing, rather than dis-esteeming, minority languages such as rTa’u; and by tolerating ‘accents’ in people who speak Tibetan as a second language, and even esteeming such individuals for their multilingual repertoires, rather than viewing such capacities as barriers to Tibetanness. These changes would go a long way to improving an ideological environment which is, to say the least, currently unsupportive of the maintenance of the rTa’u language.

37Finally, it is worth noting that many of Tibet’s minority languages, all of which are endangered to some degree, are in a similar predicament to rTa’u, facing a double bind between maintaining a language but having their Tibetanness questioned, or abandoning their language in order to claim a Tibetan identity. State structures that deny recognition to Tibet’s minority languages are unlikely to change, and thus the exclusion of these languages from major social institutions is likely to persist. Any attempts to reverse language shift in those domains is therefore unlikely to succeed. However, it is possible that a change in attitudes towards these languages by other Tibetans, and greater recognition of their social and cultural value, would help support these languages into the future, and thus help reverse the broader decline in linguistic diversity currently apparent in Tibet and the Himalaya.



Ahmad, Zahiruddhin. 1970. Sino-Tibetan Relations in the Seventeenth Century, Serie Orientale Roma XL (Roma: Istituto Italiano per il Medio ed Estremo Oriente).

Alim, H. Samy, Rickford, John, and Ball, Arnetha. 2016. Raciolinguistics: How Language Shapes our Ideas About Race (Oxford: Oxford University Press),

Calvet, Louis-Jean, trans. by Andrew Brown. 2006. Toward an Ecology of World Languages (Cambridge: Polity).

Cameron, Deborah. 1996. Verbal Hygiene (London: Routledge).

Coulthard, Glen. 2008. ‘Beyond Recognition: Indigenous Self-Determination as Prefigurative Practice’, in Lighting the Eighth Fire: The Liberation, Resurgence, and Protection of Indigenous Nations, ed. by Leanne Simpson (Winnipeg: Arbeiter Ring Publishing), pp. 187–204.

Coulthard, Glen. 2014. Red Skins, White Masks: Rejecting the Colonial Politics of Recognition (Minneapolis: University of Minnesota Press).

Cudd, Ann. 2006. Analyzing Oppression (Oxford: Oxford University Press).

Daofu xianzhi 道孚县志 [Daofu County Annals]. 1997 (Sichuan Renmin Chubanshe 四川人民出版社 [Sichuan People’s Press]).

Das, Sarat Chandra. 1902. A Tibetan-English Dictionary with Sanskrit Synonyms (Calcutta: Bengal Secretariat Book Depot).

Dobrin, Lise, Austin, Peter K., and Nathan, David. 2007. ‘Dying to Be Counted: The Commodification of Endangered Languages in Documentary Linguistics’, in Proceedings of the Conference on Language Documentation and Linguistic Theory (London: School of Oriental and African Studies, University of London),

Dorian, Nancy. 1998. ‘Western Language Ideologies and Small Language Prospects’, in Endangered Languages: Language Loss and Community Response, ed. by Lenore Grenoble and Lindsay Whaley (Cambridge: Cambridge University Press), pp. 3–12,

Edgar, J. H. 1932. ‘The Horpa of the Upper Nya or Yalung’, Journal of the West China Border Research Society 5: 69–72.

Fraser, Nancy. 2000. ‘Rethinking Recognition’, New Left Review 3: 107.

Fraser, Nancy. 2003. ‘Social Justice in the Age of Identity Politics: Redistribution, Recognition, and Participation’, in Redistribution or Recognition?: A Political-Philosophical Exchange, ed. by Nancy Fraser and Axel Honneth (London: Verso), pp. 7–109.

Gal, Susan. 2016. ‘Sociolinguistic Differentiation’, in Sociolinguistics: Theoretical Debates, ed. by Nikolas Coupland (Cambridge: Cambridge University Press), pp. 113–38.

Ganzi xianzhi 甘孜县志 [Ganzi County Annals]. 1999. (Chengdu: Sichuan Kexuejishu Chubanshe 四川科学技术出版社 [Sichuan Science and Technology Press]).

Gele 格勒. 1988. Lun Zangzu wenhua de qiyuan, xingcheng yu zhoubian minzu de guanxi 论藏族文化的起源、形成与周边民族的关系 [On the Origin of Tibetan Culture and Its Relationship with the Neighboring Ethnic Groups] (Guangzhou: Zhongshan Daxue Chubanshe 中山大学出版社 [Zhongshan University Press]).

Giraudeau, Pierre-Philippe and Goré, Françis Louis Noël. 1956. Dictionnaire français-tibétain (Tibet oriental) (Paris: Adrien-Maisonneuve).

Grillo, Ralph. 1989. Dominant Languages: Language and Hierarchy in Britain and France (Cambridge: Cambridge University Press).

Haugen, Einar. 2001. ‘The Ecology of Language’, in The Ecolinguistics Reader: Language, Ecology, and Environment, ed. by Alwin Fill and Peter Mühlhäusler (London: Continuum), pp. 57–66.

Heinrich, Patrick. 2012. The Making of Monolingual Japan: Language Ideology and Japanese Modernity (Toronto: Multilingual Matters).

Hodgson, Brian. 1853. ‘Sifan and Hórsók Vocabularies’, Journal of the Asiatic Society of Bengal 22: 121–51.

Honneth, Axel. 1995. The Struggle for Recognition (Cambridge: Polity).

Jäschke, Heinrich August. 1881. A Tibetan-English Dictionary: With Special Reference to the Prevailing Dialects (London: Kegan Paul).

Karmay, Samten G. 2005. ‘The Thirty-nine Tribes of Hor: A Historical Perspective’, in The Arrow and the Spindle: Studies in History, Myths, Rituals, and Beliefs in Tibet, Vol. 2 (Kathmandu: Mandala Book Point), pp. 181–204.

Les Missionnaires Catholiques du Thibet. 1899. Dictionnaire thibétain-latin-français (Hong Kong: Imprimerie de la Société des Missions Étrangères).

Li, Shaoming 李绍明. 1980. ‘Tangdai Xishan zhu qiang koolüe 唐代西山诸羌 考略 [Research on various Qiang groups in West Mountains during Tang Dynasty]’, Sichuan Daxue Xuebao 四川大学学报 [Sichuan University Journal] 1: 83–95.

Li, Zhiying李志英. 2015. ‘Kangqu “Baili tusi” kao 康区 “白利土司” 考 [On “Beri chieftain” in Khams]’, Zangxue Xuekan 藏学学刊13: 107–22.

Lippi-Green, Rosina. 1997. English with an Accent: Language, Ideology, and Discrimination in the United States (London: Routledge).

Makoni, Sinfree and Pennycook, Alistair. 2005. ‘Disinventing and (Re) Constituting Languages’, Critical Inquiry in Language Studies: An International Journal 2 (3): 123–57.

McBride, Cillian. 2013. Recognition (Cambridge: Polity).

Moriyasu, Takao. 1977. ‘Tibet-go siryoo tyuu ni arawareru hoppoo minzoku — DRU-GU to HOR — [A propos des tribus de l’Asie Centrale qui se trouvent dans les sources tibétaines: DRU-GU et HOR]’, Journal of Asian and African Studies 14: 1–48.

Mühlhäusler, Peter. 2006. ‘Naming Languages, Drawing Language Boundaries, and Maintaining Languages, with Special Reference to the Linguistic Situation in Papua New Guinea’, in Language Diversity in the Pacific: Endangerment and Survival, ed. by Dennis Cunningham, D. E. Ingram and Kenneth Sumbuk (Clevedon: Multilingual Matters), pp. 24–39, https://doi. org/10.21832/9781853598685-006

de Nebesky-Wojkowitze, René. 1956. Oracles and Demons of Tibet: The Cult and Iconography of the Tibetan Protective Deities (’s-Gravenhage: Mouton).

Oidtmann, Max Gordon. 2014. Between Patron and Priest: Amdo Tibet Under Qing Rule, 1792–1911 (PhD Dissertation, Harvard University),

Pauwels, Anne. 2016. Language Maintenance and Shift (Cambridge: Cambridge University Press).

Piller, Ingrid. 2016. Linguistic Diversity and Social Justice: An Introduction to Applied Sociolinguistics (Oxford: Oxford University Press).

Povinelli, Elisabeth. 2002. The Cunning of Recognition: Indigenous Alterities and the Making of Australian Multiculturalism (Durham, NC: Duke University Press).

Powers, John and Templeman, David. 2012. Historical Dictionary of Tibet (Lanham: The Scarecrow Press).

Qu, Aitang. 1996. Zangzu de yuyan he wenzi 藏族的语言和文字 [Language and Script of Tibetan] (Beijing: Zhongguo Zangxue Chubanshe 中国藏学出版社 [China Tibetology Press]).

Ran, Changsheng 冉长生. 2004. Xianshui hepan - Huoer zhi mi 鲜水河畔-霍尔之谜 [Riversides of Xianshui - The Mystery of Horpa] (Chengdu: Bashu Shushe 巴蜀 书社 [Bashu Press]).

Ren, Naiqiang 任乃强. 1981. Qiangzu yuanliu tansuo 羌族源流探索 [Exploration of the Origin and Development of Qiang People] (Chongqing: Chongqing Chubanshe 重庆出版社 [Chongqing Press]).

Robin, Françoise. 2014. ‘Streets, Slogans and Screens: New Paradigms for the Defense of the Tibetan Language’, in On the Fringes of the Harmonious Society: Tibetans and Uyghurs in Socialist China, ed. by Trine Brox and Ildikó Bellér-Hann (Copenhagen: NIAS Press), pp. 209–34.

Roche, Gerald. 2011. Nadun: Ritual and the Dynamics of Diversity in Northwest China’s Hehuang Region (PhD dissertation, Griffith University).

Roche, Gerald. 2014. ‘The Vitality of Tibet’s Minority Languages in the Twenty-first Century: Preliminary Remarks’, Multiethnica 35: 24–30.

Roche, Gerald. 2017. ‘The Transformation of Tibet’s Language Ecology in the Twenty-first Century’, The International Journal of the Sociology of Language 245: 1–35,

Roche, Gerald, and Suzuki, Hiroyuki. 2018. ‘Tibet’s Minority Languages: Diversity and Endangerment’, Modern Asian Studies 52.4: 1227–1278, https://

Roche, Gerald, and Lugyal Bum. 2018 ‘Language Revitalization of Tibetan’, in The Routledge Handbook of Language Revitalization, ed. by Leanne Hinton, Leena Huss, and Gerald Roche (London: Routledge), pp. 417–26. https://doi. org/10.4324/9781315561271

Roche, Gerald, and Yudru Tsomu. 2018. ‘Tibet’s Invisible Languages and China’s Language Endangerment Crisis: Lessons from the Gochang Language of Western Sichuan’, China Quarterly 233: 186–210. S0305741018000012

Rockhill, William Woodville. 2004 [1891]. The Land of the Lamas: Notes of a Journey Through China, Mongolia and Tibet (Delhi: Sri Satguru Publications).

Shi, Shuo 石硕, and lHa mo thar 拉毛太. 2016. ‘Tantao Zangwen wenxian zhong “hor” (huoer) de gainian ji fanwei 探讨藏文文献中 “ཧོར་(霍尔)” 的概念及范 围 [Exploring the notion and range of Hor in Tibetan documents]’, Qinghai Minzu Yanjiu 青海民族研究 [Qinghai Minzu Research] 2: 149–54.

Simpson, Audra. 2014. Mohawk Interruptus: Political Life Across the Borders of Settler States (Durham, NC: Duke University Press).

sKal-bzang ’Gyur-med 格桑居冕 and sKal-bzang dByangs-can 格桑央京. 2002. Zangyu fangyan gailun 藏语方言概论 [An Introduction to Tibetan Dialects] (Beijing: Minzu Chubanshe 民族出版社 [Minzu Press]).

Spivak, Gayatri. 2010. ‘Can the Subaltern Speak?’, in Can the Subaltern Speak? Reflections on an Idea, ed. by Rosalind Morris (New York: Columbia University Press), pp. 21–80.

Suzuki, Hiroyuki and Sonam Wangmo. 2016. ‘Lhagang Choyu: A First Look at its Sociolinguistic Status’, Studies in Asian Geolinguistics 2: 60–69.

Taylor, Charles. 1994. ‘The Politics of Recognition’, in Multiculturalism: Examining the Politics of Recognition, ed. by Amy Gutman (Princeton, NJ: Princeton University Press), pp. 25–74.

Teichman, Eric. 1922. Travels of a Consular Official in Eastern Tibet: Together with a History of the Relationship between China, Tibet, and India (Cambridge: Cambridge University Press).

Thomas, George. 1991. Linguistic Purism (New York: Longman).

Thurston, Timothy. 2015. Laughter on the Grassland: A Diachronic Study of Amdo Tibetan Comedy and the Public Intellectual in Western China (PhD dissertation, Ohio State University).

Thurston, Timothy. 2018. ‘The Purist Campaign as Metadiscursive Regime in China’s Tibet’, Inner Asia 20: 199–218,

Tournadre, Nicolas. 2005. ‘L’aire linguistique tibétaine et ses divers dialectes’, Lalies 25: 7–56.

Tournadre, Nicolas. 2014. ‘The Tibetic Languages and their Classification’, in Trans-Himalayan Linguistics: Historical and descriptive linguistics of the Himalayan area, ed. by Thomas Owen-Smith and Nathan W. Hill (Berlin: Walter de Gruyter), pp. 105–29,

Tsepon Wangchuk Deden Shakabpa. 2009. One Hundred Thousand Moons: An Advanced Political History of Tibet (Leiden: Brill), ej.9789004177321.1-1260

Tucci, Giuseppe. 1999 [1949]. Tibetan Painted Scrolls. 2 vols., 2nd ed. (Bangkok: SDI Publications).

Tunzhi (Sonam Lhundrop). 2017. ‘Language Vitality and Glottonyms in the Ethnic Corridor: The rTa’u Language’, International Journal of the Sociology of Language 245: 147–68,

Urban, Greg. 2001. Metaculture: How Culture Moves Through the World (Minneapolis: University of Minnesota Press).

Vincent, Eve. 2017. ‘Against Native Title’: Conflict and Creativity in Outback Australia (Canberra: IATSIS Press).

Wang, Stephen S. 1970–71. ‘Consonant Clusters of Tibetan Loanwords in Stau’, Monumenta Serica 29: 631–58,

Zeng, Xianjiang 曾现将. 2006. ‘Kangbei Huoer ren de laiyuan ji lishi yanbian lice 康北霍尔人的来源及历史演变蠡测 [The origin of Hor people in Northern Kham and their historical development]’, Minzu Yanjiu 民族研究 [Nationality Research] 5: 79–84.

Zeng, Xianjiang 曾现将. 2007. Hu xi minzu yu Zangyi Zoulang - yi Mengguzu wei zhongxin de lishixue kaocha 胡系民族与藏彝走廊-以蒙古族为中心的历史学考 察 [The Hu-related Ethnic Groups and the Tibetan-Yi Corridor: A Historical Investigation Based on Mongolian] (Chengdu: Sichuan Renmin Chubanshe 四川人民出版社 [Sichuan People’s Press]).


1 For instance, the township located about ten kilometers east of Luhuo County town called Srib mo (Simu) is said to be a Mongolian term, as is the village name Shwa ba thang (Xialatuo), which is said to mean ‘Yellow Plain’.

2 Ran (2004: 26–37) dates the antiquity of this settlement to approximately the fifth and sixth centuries AD.

3 Li thang in Literary Tibetan. However, local Tibetans in Li thang consider the toponym as Le thang which has the same sense as Li thang ‘grassland as a bronze mirror’. The local pronunciation follows the spelling Le thang, not Li thang.

4 Hor sras khag lnga ‘five districts ruled by the Hor prince’ in Literary Tibetan.

5 Hor dpon khag lnga ‘five Hor chieftain’s states’ in Literary Tibetan. There are also other Literary forms, such as hor khog khag lnga and hor khag sde lnga.

6 Each of them in Literary Tibetan is rTa’u, Brag ’go, Tre hor, dKar mdzes, and sTong skor. However, the components of ‘five regions’ are not consistent in literature and narratives.

7 Not giving a detailed description on the usage of hor in the western Tibetosphere within the article, we note here that in the present Ladakh, Tibetic-speakers employ hor to designate Turkic groups, especially Uyghur and Uzbek as well as hor pa for these people (Nicolas Tournadre, p. c.).

8 This collection of documents, found in the Mogao caves in contemporary Dunhuang, Gansu Province, in the People’s Republic of China, includes some of the earliest known Tibetan texts.

9 The transliteration of this document is available at

10 In the earlier period, sog denoted countries and states to the west of Tibet, mainly the present Iran as sog po stag gzi, and at present Turkey, sog po du ru ka: (Zhang ed. 1985: 2961).

11 rGya bod hor sog gyi lo rgyus nyung brjod pa byis pa ’jug pa’i ’bab stegs.

12 dBal mang paN Di ta dKon mchog rGyal mtshan (1764–1863). Although Oidtmann (2014) transcribes the first name as Belmang, we provide the local pronunciation — Welmang.

13 Zhang (1985: 3072, 3121) provides a spelling a po hor, which might follow the pronunciation of Central Tibet.

14 This group was previously named Kham-Hor in Tournadre (2005).

15 Huorong. Here and following, the Chinese pinyin for Tibetan place names are provide in the footnotes.

16 Honglong.

17 Huoqu.

18 Heni or Heranniba.

19 The following section is based on two periods of fieldwork, in 2014 and 2017. Data were collected mainly through qualitative, semi-structured interviews, with a sample of twenty in the former and ten in the latter, by way of cluster sampling.

20 Teichman then goes on to discuss the ‘curious racial mixture’ of local Tibetans, suggesting that he viewed linguistic and biological types as related.

21 A speech on Mdo Kham rTa’u area by ‘gyaur Med Tshering, 26 April 2014:; a brief talk on Mdo Kham rTa’u language by Orgyan rDorje, 12 March 2015:

22 A short take on the relationship between rTa’u language and written Tibetan by rTa’u rGyal Mtshan, 25 June 2016:

23 The Tibetan & Himalayan Library, Tibetan text archive, English definition of logs skad,

24 Suzuki and Sonam Wangmo (2016) discuss the term logs skad.

25 In addition to ‘ghost’, other suggested translations for ‘dre include ‘demon’, ‘imp’, ‘goblin’, and ‘devil’, all strongly suggestive of the term’s negative connotations (see Interestingly, Thurston (2018: 205) provides another possible interpretation for ’dre skad as ‘blended language’; the word ’dre is also a verb stem denoting ‘mix, blend.’

26 For further discussion of this system, see Chapter 5 of this volume, by Bendi Tso and Mark Turin.

27 See the poem online here,

28 Here and elsewhere, the Tibetan texts have been transcribed faithfully according to the original, even when they include ‘errors’.

29 Translation by Sonam Lhundrop.

30 In addition to the two songs described below, another extremely popular song on this theme is Three Lamas are Seated Up There (sTod gan na bla ma rnam gsum bzhugs) by Rigdzin Drolma (Rig ’dzin sgrol ma), which can be accessed at:

31 You can view Gergyal Pedma performing the song here:

32 Translation by Sonam Lhundrop.

33 See the song performed here:

34 Translation by Sonam Lhundrop.

Table des illustrations

Titre Fig. 1.1. Map of Tibet with cultural regions and prefectures. Created by the authors, CC BY.
Légende This map shows major locations mentioned in this chapter. The inset (left) shows the location of the Tibetan Plateau within the People’s Republic of China. The central map shows the Tibetan Plateau divided into three main cultural regions (Amdo, Khams, and U-Tsang), and also shows prefectures mentioned herein: 1) Nag chu; 2) dKar mdzes, and; 3) bDe chen. At right, dKar mdzes Prefecture is shown, with counties mentioned in the text: A) sDe dge; B) ’Ba’ thang; C) Li thang; D) Brag ’go; E) rTa’u, and; F) Rong brag.
Fichier image/jpeg, 33k


Ph.D. student in linguistics at La Trobe University, Australia. He is writing a descriptive grammar of the rTa’u language, a rGyalrongic language spoken in western Sichuan Province, China. He is a native of the rTa’u community and for the last decade he has been engaged in language and cultural documentation projects.

Holds a D.Litt. in linguistics from Kyoto University (2007) and is currently a post-doctoral researcher at the Department of Culture Studies and Oriental Languages, University of Oslo, Norway, and a visiting scholar at the National Museum of Ethnology, Osaka, Japan. His principal research interests are descriptive linguistics, geolinguistics, dialectology, and sociolinguistics of languages in the Tibetosphere. He has published various works on preliminary descriptions of individual Tibetic languages, grammar sketches, geolinguistic analysis, and narrative analysis with interlinear glossing. He is an author of two books: Dongfang Zangqu Zhuyuyan Yanjiu (2015) and 100 Linguistic Maps of the Swadesh Word List of Tibetic Languages From Yunnan (2018).

Senior Research Fellow in the Department of Politics, Media, and Philosophy at La Trobe University, and has previously held positions at the University of Melbourne, Uppsala University, and Qinghai Normal University. His research focuses on the politics of language endangerment and revitalization, particularly within Tibet and the Himalayas. Recent edited publications include the Routledge Handbook of Language Revitalization (with Leanne Hinton and Leena Huss) and two open access publications: Indigenous Efflorescence: Beyond Revitalization in Sapmi and Ainu Mosir (with Hiroshi Maruyama and Isa Virdi-Kroik), and Long Narrative Songs from the Mongghul of Northeast Tibet: Texts in Mongghul, Chinese, and English (with Limusishiden).


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search