Calculating the Degree of Communicative Pressure on Tibetan and Its Users: An Examination of the Situation of Lhasa Peoples’ Native Language

Yeshes Vodgsal Atshogs (Yeshe Vodsal Atsok)

Abstract: This paper proposes the concept of the “Expanded Degree of Communicative Pressure” (E-DCP) as a quantitative methodology to infer the cumulative power of the factors underpinning language variation and language change through observing the attitudes and actual language choices of language users in multilingual contexts. Using data from the China Tibetology Research Center (Zhou Wei 2003), this paper examines the E-DCP of Lhasa Tibetan to estimate the language vitality of the Tibetan language. I find that the Tibetan language in its cultural center suffers significant communicative pressure from the external language, Chinese. If this pressure is sustained over the long term, language shift will be inevitable. This situation also shows that the bilingual phenomenon in ethnic minority areas is not simply a peaceful and equal interaction between two languages but one of the processes by which an ethnic minority group loses its language.

Keywords: language attitude, degree of communicative pressure, ratio of pressure/domain, language function space, intimate and non-intimate domains, Tibetan language, Lhasa dialect

 

Introduction

The complex contact between different languages in multilingual environments can produce a range of responses including borrowing, interference, mixture, the formation of bilingualism, translingualism (see Li and Roche 2017), new language styles, language shift, or even language endangerment and extinction. These responses are determined primarily by social factors rather than linguistic ones. Ideally, if we can isolate the various social factors related to language contact and accurately evaluate the weight of these factors in language change, we should theoretically be able to predict the direction and outcome of language variation. These social factors—which include the political status of the languages and their users, economic levels, literacy, demographics, language attitudes, as well as the official language policy, the specific contexts of verbal communication, and the topics of conversation—may all impact the participants’ use of language for communication, thereby affecting the outcome of language contact and variation.

The tremendous breadth and extent of these social factors present a variety of methodological difficulties in objectively measuring and holistically evaluating the relative importance of each. While a number of methods have been developed for assessing these responses, most tend to be focused on assessing specific trends, as with, for example, the Language Vitality and Endangerment Framework developed by UNESCO’s Ad Hoc Expert Group on Endangered Languages in 2003. This paper uses the theory of “expanded communicative pressure” (Atshogs 2003) to assess the “combined force” of various social factors related to language choice. Communicative pressure is directly and explicitly expressed by the speaker’s language attitude and language choice. Methodologically, since the internal pressure is difficult to observe, we use direct observation of the users’ language choices to measure the communicative pressure.

In recent years, the introduction of Putonghua, Standard Mandarin, in Tibetan communities in the People’s Republic of China has put tremendous pressure on Tibet’s already diverse language ecology (see, for example, Roche 2017, and Roche and Suzuki 2018), leading to a variety of responses. In Amdo, a “purism” campaign has emerged (Thurston 2018), and language has featured heavily in education policy (Dak Lhagyal 2021). In order to better understand communicative pressure as well as the interaction between Putonghua and Tibetan, this paper evaluates the viability of Tibetan in Lhasa and the direction of language shift in the region by using survey data that specifically reflects the language choices .

To accomplish these goals, this paper begins with an introduction to the theory of “degree of communicative pressure” (DCP) as a method for assessing language contact and shift. Then, I propose to advance the theory with a concept of Expanded DCP (E-DCP) to account for language contact between multiple languages, and in communities with large numbers of multilingual speakers. Following this, I apply the concept to Zhou’s 2003 dataset of the Tibetan language in Lhasa, which despite being dated is among the most complete datasets examining the practical uses of Lhasa Tibetan and provides a good way of analyzing directions of language shift in Lhasa.

1.1 Language Depth Contact and Its Social Dynamic Mechanism: Degree of Communicative Pressure (DCP)

Based on an analysis of the formation of mixed languages, Atshogs in his 2003 work introduces the “Degree of Communicative Pressure” (afterward referred to as DCP) as a method for estimating the comprehensive strength of the different social factors shaping language variation and change based on the speakers’ actual language selections in different domains. Because speakers choose between two or more kinds of languages, the DCP can also be regarded as a situation in which one language faces selection pressure from some other language. The method is especially useful because it begins from actual language choices rather than social factors themselves.

The DCP method was originally designed to explain the mechanism behind the formation of mixed languages (Atshogs 2003, 2004, 2005), and it obtained good results.[1] But the existence and degree of communicative pressure from other languages are important factors in the degree and trend of language endangerment, and even in the modes and rates of extinction.

Historically, a language often moves into another language’s area to survive, which leads to immediate depth contact between two completely strange languages, and even the formation of diglossia (Sayahi 2007) and mixed language in a short period of time. Mainly due to the need to explain this sort of sudden, in-depth language contact and its outcome, the degree of communicative pressure (DCP) is defined as “the ratio of an unfamiliar language or language variety (hereafter referred to as the ‘foreign language’) that speakers are immediately required to use relative to the whole of language exchange activities in order to achieve communication” (Atshogs, 2010:381).

In his analysis of the mixed language “Dao Language”, Atshogs (2003) divided the DCP into a total of 5 levels from 0 to 4 (0 being the lowest) as follows:

Degree	Explanation
4	Participants are required to immediately adopt a certain foreign language to realize all communicative activities.
3	Participants are required to immediately face a foreign language to communicate in daily life (such as at home); but also enjoy the freedom to use the native language outside the home.
2	Participants are required to immediately face a foreign language to realize communication outside the main contexts of daily life (such as at home) but at the same time enjoy the freedom of using the native language in daily life (such as at home).
1	Participants can use both a foreign language and the native language to realize communication (as in a skilled bilingual community).
0	Participants can only use the native language without foreign language-oriented communication requirements (as in a single language community).

These five levels are assessed primarily along two dimensions: firstly, whether the communication space is complete or partial, i.e. the size and level of the space; and secondly, whether or not they must “immediately” realize communication (i.e. the time-sensitivity).

In the time dimension, the outcomes of language contact may be completely different in different temporal contexts depending on whether there is sufficient time to have contact with a “foreign language” “step by step”, or whether speakers must achieve communication in a foreign language in a very short period of time. Just like in second language acquisition (which is also a process of language contact), even when in the same language-oriented communicative space, if there is enough time, one can also successfully acquire the target language, but there tends to be incomplete acquisition, and even the possibility of interlanguage through fossilization (Selinker, 1972) if the timeline is too compressed. In “Dao Language”, we assume that the time parameter must be “immediate”, otherwise the participants have the opportunity to contact each other and acquire the other language gradually to ease the communicative pressure and lose the impetus to form a new communicative language.

At the same time, in the “space” dimension, evaluations reflect the amount of the pressure, but also, and more importantly, the different domains in which the pressure occurs[2]. Within these, the home is the important operational boundary. This is because the home is an extremely important space for human language activities, and “in many studies on multilingual behavior, the family domain has proven to be a very crucial one” (Fishman 1972: 82). At the same time, the home is often a relatively significant watershed between intimate and formal language in the whole of language communication activities. Language atrophy is often characterized by a withdrawal from the public domain into the home, with the family often becoming the last refuge of endangered languages. As David Crystal notes, “when it comes to language death, languages first become ‘internal’ languages, and finally a ‘domestic dialect’” (Language Death 18).

1.2 Ratio of Communicative Pressure: Different Language Change Trends under Different Pressure Ratios

A further complicating factor is that in situations of language contact, the pure (degree of) communicative pressure one language faces is not sufficient to fully explain the outcomes of language contact. Instead, the ratio of the communicative pressure to which the contacted languages are exposed plays a vital role in determining the outcome of language contact.[3]

The different ratios directly determine the “contact directions” of the languages, which also determines the different features of the contact. For example, Atshogs (2003 and 2004) shows how the ratio of communicative pressure determines which source language should provide new vocabulary and which should be the grammar source of the new language when two languages form a mixed language through in-depth contact. Theoretically, there are fifteen different potential ratios of communicative pressure, namely: 0:0, 0:1, 0:2, 0:3, 0:4, 4:1, 4:2, 4:3, 4:4, 3:1, 3:2, 3:3, 2:1, 2:2, 1:1. The resulting issues can generally be divided into three umbrella categories as follows:

1) When the communicative pressure is very low (1:1) for both languages, there may be language conversion. Additionally, language variation can, but does not necessarily, appear. At this point, language fusion becomes more difficult (Atshogs 2004).

2) Under conditions in which one of the two sides is under a sustained communicative pressure of between 2 and 4 degrees, and the communicative pressure on the other side is still very low, contact-induced language variation appears. To form a separate, mixed language with independent native language status, both sides in the contact should be continuously in the environment with the communicative pressure of between 3 and 4, and speaking conservatively when one side is continually under degrees of communicative pressure of 4 (such as 2:4, 3:4 and so on). The side under greater pressure often becomes the grammar provider of the new mixed language, and the other side is often the provider of vocabulary. When multiple generations live in this sort of environment, the newly created mixed language may develop into the mother tongue and obtain the status of an independent language.

(3) Other typical communicative pressure ratios (notice that DCP is ideally set as conditions where they need to achieve not ‘slow’ but ‘immediate’ communication) are as follows:

‘0:0’ is when there is no real contact, such as ‘neighbors who hear the crowing of the same rooster but never come in contact with each other.’ Language can have differentiation but no contact.

‘0:4’ is like a person has been to a foreign land and in a totally foreign language environment. The outcome is often rapid success in acquiring the foreign language.

‘4:4’ exists in situations where, for example, laborers who use different languages are suddenly concentrated and urgently need a language to communicate. In these circumstances, speakers tend to create an intermediary language.

‘2:2’ can be like ‘1:1’: at the beginning, there are small loans, and even larger scale lending; in addition, ‘2:2’ can also be a typical pidgin language environment. In the past, the Pidgin English that appeared on the Chinese coasts went something like this: the Chinese and foreigners must both face the foreign language in business, and use their native language at home.

In short, by assessing speakers’ language choices, DCP measures the amount of space and time that various languages occupy in the entirety of communicative activities in multi-language contact environments. This provides a direct way of understanding the pressure a language faces. The ratio of the pressures borne by two languages in contact, meanwhile, helps predict the features of language change.

2     Expanded Degree of Communicative Pressure (E-DCP) – Reference Index for Language Survival State in a Multilingual Environment

Despite its value in understanding the formation of mixed languages through specific language choices, the DCP is a predictive or interpretive model only under very special circumstances where the two sides in contact are completely unfamiliar and lack bilingual speakers. Therefore, in order to facilitate analysis, the DCP model also makes a series of assumptions: 1) no more than two languages are in contact; (2) the two languages are “mutually unintelligible”; 3) neither of the communities of the languages in contact have bilingual speakers; 4) language contact is required to be “immediate”, i.e. communication with the other language must be performed immediately for survival (for example, to obtain food and clothing). These conditions are rare. In order to better discuss the relationship between multiple languages in China, the calculation of communicative pressure degree needs to be adjusted accordingly. In light of this, we propose a new “expanded degree of communicative pressure” (E-DCP), defined as the proportion of communication done in languages and language varieties other than the mother tongue in the whole of a speaker’s language communication activities.

This extension of the communicative pressure must adapt to the following scenarios: 1) languages in contact need not be mutually “foreign”, and there are quite a number of bilinguals (multi-linguals), or all members are bilinguals (multi-linguals); 2) the time scale need not necessarily be “immediate;” and 3) the relevant domains need not be limited to inside and outside of the home. If conditions permit, any sufficiently complex domain space can be chosen.

Situations involving bilinguals require that we redesign the model for each degree. Again, one of the key factors for assessing communicative pressure is the space. Rather than focusing on the home, the “domain” may be analyzed at different scales; however, the most basic dichotomy is the “intimate domain” and the “non-intimate domain”. The familiar, relaxed, and casual domains are intimate, such as internal communications within the family and private conversations between friends. The public spheres of activities, such as business environments, educational contexts, religious environment, political space and other public communication spaces can be regarded as a “non-intimate domain”. Of course, upon careful study, one can find counterexamples in these areas, but for the purposes of this study, such a basic division is sufficient and has the benefit of being easy to operate.

By treating the “intimate domain” and “non-intimate domain” as the boundary, it is possible to divide the different levels of E-DCP, as follows:

0: in all the speech acts areas, there is an overwhelming use of the mother tongue;

1: in the intimate domain, the mother tongue maintains absolute superiority; in the non-intimate domain, other languages begin to appear;

2: in the intimate domain, the mother tongue is still the main language; in the non-intimate domain, other languages begin to reach the mother tongue but not surpass the mother tongue;

3: in the non-intimate domain, other languages maintain absolute superiority; in the intimate domain, other languages are used in observable proportion but do not surpass the mother tongue;

4: in the non-intimate domain, other languages maintain absolute superiority; in the intimate domain, other languages also surpass the mother tongue.

The main trend in domain spaces featuring contact between Putonghua and minority languages in China (as well as other Chinese dialects) is the gradual erosion of the spaces of a weaker language (or dialect) from the non-intimate domain to the intimate one. At the same time, when the dominant language is expanding from the non-intimate toward the intimate areas, the dominant language does not usually enter the next domain only after the previous one was fully occupied, but it needs to obtain a certain accumulation in the previous domain before it can advance to the next. The degrees of pressure, then, can be measured by the percentage of families that use “other languages” in both intimate and non-intimate domains. The proposed ratios that “other languages” occupy in the two domains are set initially as:   

0 degree: intimate domain 0, and non-intimate domain 0；

1 degree: intimate domain 0, non-intimate domain <0.25；

2 degrees:  intimate domain <0.25, non-intimate domain 0.25～0.5;

3 degrees: intimate domain 0.25～50, non-intimate domain>0.50；

4 degrees: intimate domain>0.50, non-intimate domain≤1

The distribution is as tabled below:

	Intimate Domain	Non-intimate Domain
0 degrees	0	0
1 degree	0	<0.25
2 degrees	<0.25	0.25～0.50
3 degrees	0.25～0.50	>0.50
4 degrees	>0.50	≤1

In this way, the measurement of communicative pressure is mainly a contrast between langauge choices in different domains.

After assigning these values to the languages in contact, E-DCP, then, assesses the ratio of communicative pressure in order to analyze the relationship between languages in contact and then measure the direction of variation after contact and its outcome. Logically, in the case of contact between two languages, the expanded communicative pressure ratio can also have some of the above-mentioned fifteen types of DCP ratios. Typically, they are:

“0:0”: no real contact, such as “neighbors never in contact with each other”.

“2:3, 2:4, 1:4, 0:4”: in the competition of two languages, one language (pressures are 2, 1, 0) has the upper hand and continues, while the other language (pressure is 4) is extremely endangered.

But for the scenario envisaged by E-DCP, 2:2, 3:3, and 4:4 are too special to appear.

In theory, E-DCP can be used for any language (or dialect) which is in contact with others to assess the relationships of their mutual communicative pressure. For example, E-DCP can be applied in the evaluation of the integral relationship between languages, such as the overall pressure relationship between Tibetan and Chinese, or the overall assessment of the global relationship between Chinese and English; the evaluation can also be partial, such as the communicative pressure relationship between Chinese and English in China; it can also be used in a very specific geographical location, for example, the communicative pressure relationship between Tibetan and Chinese in Lhasa, or in Shanghai. And it can even be applied in a much smaller specific communication environment, such as a foreign language learning class, in which the influences of different “pressure ratios” on language learners are analyzed. At present, the more practical significance of the application is in language contact and endangerment research, where E-DCP provides an objective and concise method for obtaining data and calculating analysis. To illustrate, this paper applies this E-DCP to the Tibetan spoken in Lhasa.

3  Analysis of the Vitality of Tibetan –A Case Study of Lhasa City

Due to the complexity and sensitivity involved in studying Tibet, as well as the difficulty of obtaining first-hand information about language practice, we use data from Zhou Wei周炜(2003) as a stand-in to measure the dynamics of language choice in Lhasa. This work reports about investigations into language use in Tibetan society, and is considered “the first book to study the bilingual phenomenon and language usage in Tibetan society from the perspective of Sociology of Language” (16). Although the study obtains information via questionnaire rather than empirical observation of actual language use,[4] and although the information included is somewhat dated, based on a door-to-door survey of 138 households, this work provides some of the most detailed demographic information we have about Tibetan residents in Lhasa and the surrounding villages and communities (ibid pp.119-20). We primarily use the following two sets of data.[5]

Table 1: Language Choices of the Heads of Households in Public Places

Source: 周炜 [Zhou Wei]. 西藏的语言与社会 [The Languages and Society in Tibet]. 中国藏学出版社 [China Tibetology Publishing House], 2003, table 8. P. 128.

Domains Languages	Government Office		School		In Shops		Sweet Tea House		Business Street		Hospital		Meeting		Between Friends
	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%
Tibetan	34	25	27	20	19	14	78	*56	22	16	19	14	30	22	72	52
Chinese	24	17	26	19	28	20	5	4	26	19	30	22	47	34	5	4
Bilingual	80	58	85	*61	91	*66	55	40	90	65	89	*64	61	44	61	44
Total	138	100	138	100	138	100	138	100	138	100	138	100	138	100	138	100

Table 2: The Language Usages of the Family Members in Eight Family Occasions

Source: Ibid. P.130

Domains Languages	Family Together		Between Kids		At the Table		Entertainment Time		Between Couples		Chat Time		Housework Time		Playing Kids
	Number of People	%	Number of People	%	Number of People	%	Number of People	%	Number of People	%	Number of People	%	Number of People	%	Number of People	%
Tibetan	89	65	72	52	95	69	70	51	93	67	78	57	98	71	67	*49
Chinese	2	1	2	1	1	1	4	*2	1	1	3	2	2	1	1	*1
Bilingual	47	34	64	*47	42	30	64	47	44	32	57	41	38	28	70	*51
Total	138	100	138	100	138	100	138	100	138	100	138	100	138	100	138	*100

In order to make these data legible for evaluating degrees of communicative pressure, these must be processed along the following lines. Firstly, there are some obvious problems with the data calculation, and some have been corrected and expressed in bold (such as in table 10, “The Family Once” [“全家以前”] being replaced by “Family Together” [“全家一起”]). The corrections to the original table are marked with “*” (10, in total). They are mainly rounding errors.

Secondly, since the title of table 10 uses the statement “family members…language use,” it seems that the object of the survey is different from table 8. Table 8 focuses on “Heads of Households” (“户主”) while “Family Members” (“家庭成员”) are the focus of table 10. However, because the total number of people is 138 for both tables, it is inferred that the data is obtained from the same questionnaire completed by the heads of the households. Therefore, we use the materials in the two tables together comparatively, so that the “Number of People” (“人数”) should more precisely be the “Number of Households” (“户数”) represented by their respective “Heads of Households” (“户主”).

Additionally, the classification of 8 family circumstances in table 10 is far-fetched, since “Family Together” (“全家一起”) and “At Mealtimes” (“吃饭时”), or “Entertainment Time” (“娱乐时”) or “When Chatting” (“闲谈时”) all might mutually overlap. So too with “Between Husband and Wife” (“夫妻间”) and “When Chatting,” while it makes little sense to place “Between Kids” (“孩子间”) and “Kids Playing” (“孩子玩耍”). Fortunately, a much more detailed classification of the family circumstances is not needed here, but a whole grasp of language usage in an “intimate space” is necessary. So, the comprehensive calculation of the statistics of individual items to grasp the overall outlook is possible.

Similarly, in table 8, “In Shops” (“在商店”) and “Business Street” (“商业街”) are two easily confused concepts. And statistically they are also very similar, so they can be merged into “In Shops.” So the category of “Public Spaces” (“公共场所”) can also be a bit more streamlined. By contrast, “Government Office” (“在机关”) and “during meetings” (“开会时”) may also have a certain degree of overlap, but the survey data of the two items is very different; therefore, they will be retained in separate columns. It is also a pity that there is no column for “places of worship.” Perhaps this is due to the fact that it is a sensitive question, or because religious activities occupy a small portion of their lives, especially for “cadres” (“干部”), “staffs” (“职工”), and “students” (“学生”). Beyond this, all the domains for an ordinary citizen involving education, administration, hospital, business, and other fields are included.

Fourthly, one important point of note from this survey is the inclusion of a “bilingual(双语)” response. As opposed to a linguistic definition of bilingualism, this is best understood as meaning “in some occasions, it is possible to use Chinese, Tibetan, or both Chinese and Tibetan alternately”. Therefore, it is more appropriate to name the data item as “Both”. The corresponding “Tibetan” option may be understood as “in certain occasions, only Tibetan is used”, and the “Chinese” option is understood as “in certain occasions, only Chinese is used”.

After processing the data in the manner described above, we get the following data:

Table 3

Domains Languages	Government Office		School		Shop		Sweet Tea House		Hospital		Meeting		Between Friends		Within Family
	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%
Tibetan	34	24.64	27	19.57	41	14.86	78	56.52	19	13.77	30	21.74	72	52.17	662	59.96
Chinese	24	17.39	26	18.84	54	19.57	5	3.62	30	21.74	47	34.06	5	3.62	16	1.45
Both	80	57.97	85	61.59	181	66.58	55	39.86	89	64.49	61	44.20	61	44.20	426	38.59
Total	138	100	138	100	276	100	138	100	138	100	138	100	138	100	1104	100

Finally, the statistical data about “family” and “public places” needs to be changed to the “intimate domain” and “non-intimate domain.” This is primarily achieved through combining “Sweet Tea House” and “Between Friends” from public places and combining them with the data from table 10 to form the “intimate domain,” while the remaining “public places” can constitute the “non-intimate domain.” After all, it is no problem to classify activities involving friends in the “intimate domain.” Meanwhile, anyone familiar with the Lhasa “Sweet Tea House” knows that it is a very popular, simple, casual street space, in which people can communicate with each other without consideration of identity. So, it is an ideal place for communication between friends. Processed in this way, we get the following table:

Table 4

Domains Languages	Non-intimate										Intimate
	Hospital		Shop		School		Meeting		Government Office		Between Friends		Sweet Tea House		Within Family
	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%
Tibetan	19	13.77	41	14.86	27	19.57	30	21.74	34	24.64	72	52.17	78	56.52	662	59.96
Chinese	30	21.74	54	19.57	26	18.84	47	34.06	24	17.39	5	3.62	5	3.62	16	1.45
Both	89	64.49	181	66.58	85	61.59	61	44.20	80	57.97	61	44.20	55	39.86	426	38.59
Total	138	100	276	100	138	100	138	100	138	100	138	100	138	100	1104	100

Next, after combining all the data into the umbrella categories of intimate and non-intimate domains, we get the following results:

Table 3

Domains Languages	Non-intimate		Intimate
	Number of Households	%	Number of Households	%
Tibetan	151	18.24	812	58.84
Chinese	181	21.86	26	1.88
Both	496	59.90	542	39.28
Total	828	100	1380	100

The classification of “Tibetan”, “Chinese”, and “both” can also be analyzed in terms of “whether Tibetan must be used”, so table 3 can be sorted as table 4, in which “Chinese” and “both” are combined:

Table 5

Domains Languages	Non-intimate		Intimate
	Number of Households	%	Number of Households	%
Must Use Tibetan	151	18.24	812	58.84
Tibetan Might Not Be Used	677	81.76	568	41.16
Total	828	100	1380	100

From this calculation, the percentage of Lhasa residents who do not use Tibetan, then, is greater than 25% but less than 50% in the intimate domain; while in the non-intimate domain, it is 81%, exceeding 50%. According to the operational definition of E-DCP, as early as 1999, the communicative pressure on the Tibetan people of Lhasa had reached level 3, meaning that other languages maintain absolute superiority in the non-intimate domain while also being used in observable proportion (though without surpassing the mother tongue) in the intimate domain.

Of course, logically speaking, the “Both” category can actually be separated into “Tibetan” and “Chinese.” If we assume that these can be split evenly, then the statistics of the “Both” category are divided into two parts, and in consequence, they are added to “Tibetan” and “Chinese” respectively. The results are as follows:

Table 6

Domains Languages	Non-intimate										Intimate
	Meeting		Hospital		Shop		School		Government Office		Between Friends		Sweet Tea House		Within Family
	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%	Number of house- holds	%
Tibetan	60.5	43.84	63.5	46.01	131.5	47.64	69.5	50.36	74	53.62	102.5	74.28	105.5	76.45	875	79.26
Chinese	77.5	56.16	74.5	53.99	144.5	52.36	68.5	49.64	64	46.38	35.5	25.72	32.5	23.55	229	20.74
Total	138	100	138	100	276	100	138	100	138	100	138	100	138	100	1104	100

The analysis based on intimate vs non-intimate domains is as follows:

Table 7

Domains Languages	Non-intimate		Intimate
	Number of Households	%	Number of Households	%
Tibetan	399	48.19	1083	78.48
Chinese	429	51.81	297	21.52
Total	828	100	1380	100

From that point of view, at the time the data was collected, the Chinese language had just surpassed Tibetan in the non-intimate domain and reached 51.8%, while in the intimate domain Chinese had not yet reached 25%, but at 21.52% the number is very close to 25%. It can therefore be said that at the end of the twentieth century, the native language of the people of Lhasa bore level 3 social pressure, or was developing from 2 to 3.

If we then assume that the Tibetan residents of Lhasa were largely monolingual and used Tibetan almost exclusively in both intimate and non-intimate domains prior to the establishment of the People’s Republic of China (though with the recognition that Lhasa was not entirely homogenous, with the presence of both Hui Muslim and Han traders, Manchu ambans, etc.), then there has been a diachronic change with each language’s presence in Tibetophone communities in Lhasa between the 1950s and 1990s. It also represents the change of communicative pressure borne by Lhasa residents and the developing trend from 0 to 3 over this period.

The Ratio of Domains or Ratio of Communicative Pressure: Observations on the Tibetan and the Chinese Languages in Lhasa

The data presented above, however, only reflects the language choices of Tibetan families in Lhasa, and yet, Lhasa City also has a substantial Han population. According to the statistical yearbook of the Tibet Autonomous Region, the TAR had a Han population of 123,558 in 2008, almost two times the population (65,101) in 1991. Usually, nearly half of the TAR’s Han population live in Lhasa City’s Chengguan District. For the correspondence with other data in this paper, based on the 1990 census data used by Zhou Wei, the population of the Han nationality in the Chengguan District of Lhasa City is more than 40,000 people, accounting for 28.9% of the total population in the city’s Chengguan District (205). This does not include the temporary population that actually comprises more than half of the resident population (Zhou 71).

As early as 1998, 82.2% of the Tibetan ethnic urban population in Tibet had mastered both Tibetan and Chinese (Zhou Wei, 2003: 438). In other words, most young Tibetans with whom Han people often need to communicate are able to communicate in Chinese. As a result, very few of the Han Chinese residents in Lhasa master Tibetan, and they primarily use Chinese even when communicating directly with Tibetan native speakers. Therefore, it can be said that the communicative pressure on the native speakers of Chinese in Lhasa’s urban areas can be evaluated as 0. So, in Lhasa, the ratio of communicative pressure between Tibetan and Chinese is 2:0 or 3:0.

Readers may doubt the significance of a communicative pressure ratio of 2:0 or 3:0 in a city in which the Tibetan population remains the absolute majority. To this we must say that the communicative pressure is measured through actual language choice. In this way, a small number of Chinese native speakers can use Chinese in all the language domains without the pressure of using other languages for survival despite living in a community with a vast majority of Tibetan native speakers. This reflects the immense sociopolitical power of the Chinese language. The power of Chinese, of course, comes not only from the tens of thousands of Chinese native speakers in Lhasa City, but from the composite force of the Chinese language’s presence in politics, the economy, and culture as well. This force is sufficiently powerful to enable a small number of Chinese people to enter a linguistically foreign area and bear little communicative pressure.

Bilinguals: A Link in the Loss of the Mother Tongue

For a “typical,” “ideal,” “bilingual” community, perhaps people can think abstractly that it is a state between two monolingual communities. Different from the monolingual societies on either end of the spectrum, people from the two language groups get along with each other closely in this state, each mutually mastering the other’s language, freely using both languages in the various communicative fields and occasions, and creating the ideal harmonious bilingual society. But this is clearly not the case in Lhasa. One fundamental difference is that the “bilingual” phenomenon just belongs to one group and not both. That is to say, “bilinguals” and “bilingual phenomena” are limited to the native speakers of minority languages, rather than “both” ethnic groups. Most of the people who master the other side’s language are members of minority ethnic groups. Native speakers of the common language (in this case, Chinese) are usually monolinguals, and few of them master minority languages.

Therefore, the “bilingual” phenomenon in minority areas is the result of the common language eating into the functional space of the minority language; bilingualism is not a balance or compromise between the two languages, but the process of the native speakers of minority languages surrendering their own languages’ functional space. When this state continues, the community’s language practices go from monolingual to bilingual to monolingual, and the second monolingual language is just the shifted language after the loss of the mother tongue. Seen in this light, this “bilingual” phenomenon is nothing but a stage of losing the mother tongue. In fact, this kind of “bilingual” phenomenon not only describes the survival state of minority nationalities’ languages but also depicts the survival state of weak languages around the world besides the survival state of the Chinese dialects under the popularity of the common language.

Disaggregating language attitudes and choices along generational lines further demonstrates this shift toward bilingualism. In fact, since the beginning of the twenty-first century, the Lhasa Tibetan language has not only been under growing communicative pressure, but quite a number of young people have abandoned their native Tibetan language, and new ethnic Tibetan monolingual Chinese speakers have emerged like bamboo shoots after a spring rain. For example, Zhou Wei’s survey data also suggest that by 1999 some young Tibetans living in Lhasa had already completely shifted to the Chinese language. According to a survey of Tibetan senior high school students in Lhasa, 12% of the students first spoke the Chinese language, with 10% speaking Mandarin, and the remaining 2% speaking Chinese dialects (Zhou 169).[6]

Additionally, Chinese language competence is on the rise from the older generations to younger ones. For many, the Chinese language competence of the new generation of urban Tibetans has surpassed their Tibetan ability. According to Zhou Wei (2003:124, table 3), the data on the language ability of the principal members of the family interviewed are as follows (and plotted in Figure 4): 

Additionally, in a survey about the “sense of language” among Tibetan students, 67% of high school students hope that Mandarin can be used as the language of instruction in primary and middle schools; for those who hope to use Mandarin in teaching the science curriculum, the number increases to 84%. While slightly less than 68.1% of middle school students supported Mandarin as the teaching language, for primary school students the number is as high as 85.48% (Zhou 178-179). In fact, in middle and high schools, where Mandarin is already the language of instruction in 94% of maths and science classrooms (Zhou 172), 75% of Tibetan high school students already prefer Mandarin in new media such as television, radio, movies, books, newspapers, magazines, and songs. For junior middle school students, the number is slightly lower than 56%, while for primary school students it rises to 76% (Zhou 177). Together, these data demonstrate the generational shift toward bilingualism and toward Putonghua currently underway in Lhasa.

Conclusion

In this article, we have developed the idea of Expanded Degrees of Communicative Pressure in order to assess the social pressure exerted on languages in contact. In applying this concept to data on language choice in different domains and functional spaces as reported by Tibetans from Lhasa, we have come to recognize that the Tibetan language has suffered considerable communicative pressure from external languages in its cultural center of Lhasa, with an estimated pressure between 2 and 3 degrees at the end of the 20^th century. Simultaneously, the contacted language (Chinese) has nearly no pressure. If the pressure on the Tibetan language in Lhasa persists, there will inevitably be a gradual language shift. Although we have had to process, and make certain assumptions with, this extant dataset, the data clearly demonstrate that the status of Tibetan has changed sharply with the increase in the bilingual population, and a group among the younger generation has even abandoned their native language entirely. Once the communicative pressure fully develops to the 3^rd degree, language shift will be quick and inevitable. Unfortunately with education policy across the Tibetan cultural region increasingly emphasizing instruction in Putonghua, and the increasing presence of the Chinese language media in Tibetan spaces, the pressure on the Tibetan language will only increase in the coming years.

Future research, meanwhile, would do well to examine minority Tibetan languages (see Roche and Suzuki 2018), re-examine the situation with more recent datasets, and include information on language choice in religious and ritual domains (which may persist even after language shift in other domains), and examine data from beyond the Tibetan Autonomous Region.  This is especially true in places like Qinghai Province where the Tibetan language has been maintained very well. In Qinghai, local governments prepared an aggressive “Qinghai Province Long-term Educational Reform and Development Plan (2010-2020),” intended to implement a plan that would comprehensively take Chinese as the language of instruction from cities to the grasslands and in basic education, including one year of preschool, all over a five-year period. The implementation of the plan would have increased the communicative pressure on the ethnic languages, quickly endangering them. This “plan” met with considerable resistance from Tibetan communities and had not yet been implemented at the time of writing. The government of Hainan Autonomous Prefecture, Qinghai Province, began promoting a similar plan in early 2017, which led to a massive social media backlash (see Dak Lhagyal 2021). Only time will tell how things will change in Lhasa and beyond, and what local communities can do in the face of this continuing communicative pressure.

  

Works Cited

Atshogs, Yeshes Vodgsal [意西微萨·阿错]. 藏、汉语言在“倒话”中的混合与语言深度接触研究 (Research on Mixing of Tibetan and Chinese in Daohua & Relative Languages Deep-Contact Study). 2003. 南开大学博士学位论文 [Nankai U, Doctoral Dissertation].

—. 倒话研究 [Studies of Daohua Language]. 民族出版社 [Ethnic Publishing House], 2004.

—. “语言深度接触机制与藏汉语言类型差异问题” [“The Study of Language Deep-Contacting Mechanism and the Typological Difference between Tibetan and Chinese”]. Journal of Chinese Linguistics (USA), vol. 33, no. 1, Jan. 2005, pp. 1-33.

—. “交际压力度（DCP）与混合语形成机制—以倒话为例” [“Degree of Communicative Pressure and the Formative Mechanism of Mixed Language, with Example of Daohua”]. 研究之乐：庆祝王士元先生七十五寿辰学术论文集 [The Joy of Research II: A Festschrift in Honor of Professor William S-Y. Wang on His Seventy-fifth Birthday], 载潘悟云、沈中伟主编 [co-edited by Pan Wuyun and Shen Zhongwei], 上海教育出版社 [Shanghai Educational Publishing House], 2010, pp. 376-388.

Crystal, David. “Endangered Languages: What Should We Do Now?” Language Documentation and Description, edited by Peter K. Austin, vol. 1, 2003, pp. 18-34.

—. Language Death. Cambridge UP, 2000.

Fishman, Joshua A. “The Relationship between Micro- and Macro- Sociolinguistics in the Study of Who Speaks What Language to Whom and When.” Sociolinguistics, edited by J. B. Pride and Janet Holmes, Penguin, 1972, pp. 15-32.

—. “Sociolinguistic Perspective on the Study of Bilingualism.” Linguistics, vol. 6, no. 39, 1968, pp. 21-49.

胡明扬 [Hu Mingyang]. “混合语理论的重大突破——读意西微萨·阿错著《倒话研究》” [“A Major Breakthrough in Hybrid Language Theory: A Review of Studies of Daohua Language by Yeshes Vodgsal Atshogs”]. 中国语文 [Chinese Language], vol. 2, 2006, pp. 187-190.

Lhagyal, Dak. “‘Linguistic Authority’ in State-Society Interaction: Cultural Politics of Tibetan Education in China.” Discourse: Studies in the Cultural Politics of Education, vol. 42, no. 3, 2021, pp. 353-367.

马戎 [Ma Rong]. “Preface.” 西藏的语言与社会 [The Languages and Society in Tibet]. 中国藏学出版社 [China Tibetology Publishing House], 2003. 

Minett, J.W. and W.S.-Y. Wang. “Modelling Endangered Languages: the Effects of Bilingualism and Social Structure.” Lingua, vol. 118, no. 1, 2008, pp. 19-45.

Mufwene, Salikoko S. The Ecology of Language Evolution. Cambridge UP, 2001.

国立国語研究所編 [National Language Research Institute, editor.] “待遇表現の実態：松江24時間調査資料から”  [“Some Aspects of Honorific Expressions: in Special Reference to Discourse”]. 国立国語研究所報告41 (The Report of the National Language Research Institute, No. 41). 国立国語研究所[National Language Research Institute], 1971. 

Roche, Gerald. “Introduction: The Transformation of Tibet’s Language Ecology in the Twenty-first Century.” International Journal of the Sociology of Language, no. 245, 2017, pp. 1-35.

Roche, Gerald and Hiroyuko Suzuki. “Tibet’s Minority Languages: Diversity and Endangerment.” Modern Asian Studies, vol. 52, no. 4, 2018, pp. 1227-1278.

Sayahi, Lotfi. “Diglossia and Contact-induced Language Change.” International Journal of Multilingualism, vol. 4, no. 1, 2007, pp. 38-51.

Selinker, L.  “Interlanguage.” International Review of Applied Linguistics in Language Teaching, vol. 10, no. 3, 1972, pp. 209-231.

Thurston, Timothy. “The Purist Campaign as Metadiscursive Regime in China’s Tibet.” Inner Asia, vol. 20, no. 2, Oct. 2018, pp. 199-218.

UNESCO Ad Hoc Expert Group on Endangered Languages. “Language Vitality and Endangerment.” UNESCO, Mar. 2003. www.unesdoc.unesco.org/images/0018/001836/183699E.pdf.

Urla, Jacqueline and Crista Burdick. “Counting Matters: Quantifying the Vitality and Value of Basque.” International Journal of the Sociology of Language, no. 252, 2018, pp. 73-96.

周炜 [Zhou Wei]. 西藏的语言与社会 [The Languages and Society in Tibet]. 中国藏学出版社 [China Tibetology Publishing House], 2003.

Notes

[1] Logically, this method can also be applied in the fields of second language acquisition and language endangerment (Hu Mingyang 2006). The amount of degree of communicative pressure in learning greatly influences the process and outcome of second language acquisition.

[2] The actual factors involved in a domain are very complex, as communicative topics, role relationships, and communicative context and so many other factors are related to this (Fishman 1972). But when this article uses “domain” to measure communicative pressure, the domain level needed to be considered is actually quite simple, only needing analysis inside and outside the family or the following big classes of “intimate domain” and “non-intimate domain”, so not considering the finer factors regarding the impact of the domain will not greatly affect the analysis.

[3] Atshogs (2003, 2004) used a similar analytical method but did not explicitly sum it up into corresponding categories.

[4] Urla and Burdick (2018) and others have demonstrated that such quantitative data may provide an aura of objectivity while obscuring more complex situations.

[5] Zhou Wei (2003)  reported questionnaire survey results respectively using the terms “urban Tibetan residents” (“城镇藏族居民”), “Tibetan cadres” (“藏族干部”), “Tibetan business personnel” (“藏族工商人员”), and “Tibetan students” (“藏族学生”). However, these categories are not of the same level and not exclusive, so we just use survey data of “urban Tibetan residents” here because this survey category is relatively the most representative; urban Tibetan residents include cadres, technicians, workers, businessmen, farmers, the unemployed, and individuals with other occupations, and the levels of education range from illiteracy to primary school to university.

[6] In Zhou Wei’s survey of Tibetan junior middle school students, the proportion of students who first acquired Chinese seems to be only 4%, but in fact, it can be seen that 11% of junior middle school students are not skilled in speaking Lhasa Tibetan (160). (The survey mentions that there are two students from outside Lhasa; if these two students are included in the 11%, after excluding them, non-fluent speakers have reached 10.5%). For the data “first acquired which language,” “Chinese” seems to be only 2%, but this group of data on the primary school itself has a big problem that makes it difficult to use. The content of the data is the following: Lhasa 95%, Tibetan 3%, Mandarin 2%, Chinese dialects 0%, and Tibetan-Chinese bilingual 19%. (Zhou 146). From the sense of “first,” the items should be exclusive and cannot be stacked, but the total of these five items is 119%; if only including the first four items, the total is just 100%, but the fifth item “first acquired Tibetan-Chinese bilingual” is hard to understand. Perhaps the first four groups are further divisions of monolinguals. If so, the Tibetan primary school students who first acquired Chinese simultaneously with Tibetan have reached 19%, which is a staggering figure.

 

Yeshes Vodgsal Atshogs (Yeshe Vodsal Atsok, །ཡེ་ཤེས་འོད་གསལ་ཨ་ཚོགས། 意西微萨•阿错) is Professor and Director of Sino-Tibetan Languages Research Center of Nankai University, Tianjin, China.