Losing Chinese as the First Language in Thailand

Much of the classical sociolinguistic literature on the issue of language maintenance amongst Sino-Siamese communities rely on language-use data on a sample of third-fifth generation Thais with Chinese ethnic roots, the majority of whose great grandparents immigrated to Thailand during the nineteenth and early decades of the twentieth century. As such, in this literature, are excluded the large sample of first-generation overseas Chinese and second-fifth generation ethnic Chinese immigrated to Thailand more roughly between 1970 and 2000. Thus, the linkage between factors governing language maintenance are to be interpreted with due caution in view of limitations in the samples discussed in the extant literature. The current study includes an assessment of the language-usage patterns and present perceptions of the vitality and sustainability of Chinese variants amongst not only the older generations, but also amongst those who have more recently immigrated, thereby rounding out a much fuller picture of the current sociolinguistic situation of those of Chinese descent domiciled in Thailand. Additionally, this study aims to examine the full range of sociological factors that contribute to the maintenance and extensive shift away from Chinese variants toward Thai and Mandarin. It accordingly complements the earlier research of Morita, but not lending support to the Skinnerian paradigm of assimilation. The sample gathered consisted of eight dialect groups. Three types of instruments of research-a questionnaire, observations, and semi-structured interviews-were utilized in collecting data. The data suggest that language-usage patterns, perceptions of vitality and sustainability of ethnic languages are strongly linked to generational changes in ethnic-linguistic identities and discursive practices governing social inclusion/exclusion, amongst other factors.


Introduce the Problem
This contribution to language-use, language-vitality and language-shift research was carried out in the Bangkok Metropolis area of Thailand.In today's increasingly monolingual Standard Thai-speaking population in the Sino-Siamese communities discussed, the increasingly weakening vitality of Chinese variants spoken as first languages (henceforth L1s) and/or ancestral/ethnic/heritage languages (henceforth HLs) spoken by members of the communities under consideration and the shift away from Chinese variants toward the habitual use of the official state language-Standard Thai-and Mandarin are two notable sociological phenomena.
As a complement to previous sociolinguistic data, this article examines three interrelated social phenomena-language use, vitality and shift-which underline the dynamic interaction between these phenomena.In a summary of previous studies, three particular issues are identified: Previous sociolinguistic research investigations of ethnic Chinese communities in Thailand are not current (Skinner, 1957(Skinner, , 1958(Skinner, , 1963(Skinner, , 1973;;Boonsanong, 1971;Chan & Tong, 1993;Hill, 1998;Morita, 2003Morita, , 2007)).As such, these investigations were largely restricted to the sociolinguistic experiences of third-fifth generation ethnic Chinese (their great grandparents migrated to Thailand during the nineteenth century and early decades of the twentieth).There has therefore been an extreme paucity of studies investigating first-third generation overseas Chinese and later generations of ethnic Chinese who have more recently immigrated in the three decade 1970-2000 period.Furthermore, the disparity with regard to language-usage patterns (influenced by intra-and inter-ethnic interactions across communicative domains) amongst the old and the new Chinese diaspora also warrants further investigation.
Additionally, previous studies (e.g., Skinner, 1957;Morita, 2003Morita, , 2007) ) overlooked the disparity shown by different Chinese immigrant subgroups-dialect groups-in the degree of ethnic language maintenance (henceforth LM) and language shift (henceforth LS).Previous research seemed to concur in viewing Teochew speakers as constituting the largest ethnic Chinese community in Thailand, in as much as the 56% of Teochew speakers outnumbered the 16% Hakka, 11% Hainan, 7% Cantonese and 7% Hokkien speakers reported by Smalley (1994).The Teochew variety was viewed as the commercial lingua franca amongst the elderly members of Chinese business circles in China Town (Yaowarat).Recalling the distinction between dominant and subsidiary groups, most scholars and members of the general public tended to accept the stereotypical or received view of Teochew as the dominant group as juxtaposed to the subsidiary groups of those of Chinese descent or partial Chinese descent, including speakers of Cantonese, Hakka, Hainanese, amongst others.It thus seems clear that the Teochew variant is believed to be well-preserved.
Moreover, sociolinguists debated and provided broad documentation for the root causes of LMLS from a variety of points of view.It should be pointed out that such researchers as Fishman (1991Fishman ( , 2001) ) and Mukherjee and David (2011), amongst others, attempted to understand the various sources of sociological effects in determining the phenomena of LMLS.One avenue of research on the reasons for LS amongst Sino-Siamese societies was the approach adopted by Morita (2003Morita ( , 2007)).She was a pioneer in examining factors contributing to the shift away from Chinese variants toward Thai (not from Chinese variants toward Mandarin), including such factors as (1) the decline of Chinese heritage language education (this factor is highly contested); (2) pro-Thai/assimilation policies; and (3) positive attitudes toward Thai society (embracing the Thai ethnic identity).However, she did not address the extent of LM.In addition to the just mentioned causes, there are yet other key factors of equal importance or more crucial to the promotion of LMLS for the communities in question.

Purpose of the Study
In response to the aforementioned issues stemming from previous research, this article has multiple purposes in view of three interconnected objectives: 1) seeks to address the extent of the intergenerational LS from various Chinese variants to Standard Thai and Mandarin, by means of examining intra-and inter-ethnic language-usage patterns across communicative domains, utilizing a language-use questionnaire; 2) aims to arrive at a better understanding of the present status of perceptions of the vitality/sustainability of Chinese variants, using (subjective) vitality perceptions (55 items) and the Expanded Graded Intergeneration Disruption Scale (henceforth EGIDS); and 3) investigates reasons for the LM of both bilingualism and diglossic code-mixing, as well as sociological factors affecting the significant LS toward the habitual use of Standard Thai and Mandarin.

Sino-Siamese Community: A Brief Overview
This section provides background and terminology for the study, while focusing on the specification of the geographical distribution of Chinese dialects, demographic information and language profiles of various Chinese dialect groups represented in Bangkok, and definitions of various ethnic Chinese identities.The great Chinese diaspora commenced in the 19th century when disparate waves from Fujian, Guangdong and Yunnan Provinces in Mainland China migrated to Thailand in accordance with different timescales.An estimated 14% of Thailand's total population of 66,720,153 (July 2011 est.) 19th-20th century (1918-1933) Coolie/poor, illiterate, unskilled, landless agricultural laborers/peasants Large-scale of immigration of Chinese to Thailand did not commence until the 1800s: Increase to 450,000 (1882-1917) 500,000 (1918-1931) The greatest n=15,460 in 1927 Decreased to 250,000 (1932-1955) After 1949 (1950's-1960's) More educated and literate class   2010-2013;Premsrirat, 2007;Smalley, 1994; definitions of terminology in reference to Thai Chinese (three categories): (1) Thais of Chinese descent; (2) ethnic Chinese; and (3) overseas Chinese) are derived from Chansiri (2007) Note.The reason why Southern-Min Taiwan participants are grouped (group 6) with Kinmen participants lies in the fact that their homelands are in geographical proximity and the two linguistic variants in question are more mutually intelligible than other speech variants.Similarly, Cantonese and Hong Kong immigrants are grouped together (group 1), in as much as they all speak Cantonese as their HL.

Language Demographics Then and Now: Chinese Dialects as Threatened Heritage Languages in Thailand
In view of LS, one may wonder how threatened are Chinese variants spoken as L1s and/or HLs in Bangkok?Multilingual practices, language contact and code-mixing were common phenomena in early modern times, particularly in urban centers such as Bangkok which attracted migrants.Most ethnic Chinese and part-Chinese families reported that a HL was spoken at home.Chinese variants were also used for intra-ethnic verbal communication in public domains.Nonetheless, Chinese variants today face an uncertain future.The present-day Bangkok (including numerous Sino-Siamese communities) is commonly depicted as monolingual entities juxtaposed with monolingual Thai-speaking Thais with ethnic Chinese roots.The Sino-Siamese communities in Bangkok have a large and growing number of non-Chinese speaking members.Fewer and fewer Sino-Siamese children grow up as L1 speakers of their HLs.In studying numerous Chinese and part-Chinese families, it was found that Thai was the only language spoken at home and in public domains.

Language Shift
In our post-industrial world, LS is a common phenomenon in most developing and developed societies.In conceptualizing 'language shift,' the researcher follows the view of social dialectologists/sociolinguists and linguistic anthropologists/ethnographers (e.g., Gal, 1978;Smith-Hefner, 2009;Mukherjee & David, 2011).The extensive body of LS research has been driven by the theory that some language-minority individuals and/or groups are constantly seeking opportunities for upper social mobility, a necessary condition of which is very often LS.This previous research shows that LS is the end result of a single individual or a speech community/group, consciously or unconsciously, for a multitude of reasons eventually adopting a dominant local language in lieu of their mother tongues and/or HLs.In this scenario, the mother tongue and/or HL usually become lost.Both historically and currently, there have been substantial efforts and rigorous research to capture and document the key variables and contextual factors that collectively contribute to LMLS, language loss, reverse language shift (henceforth RLS) (Fishman, 1991(Fishman, , 2001) ) and language revitalization in efforts at restoring ethnolinguistic vitality (henceforth EV) (proposed by Giles, Bourhis, & Taylor, 1997; also see Yagmur & Ehala, 2011).(For a fuller review of the definition of the key concept 'language vitality' proposed by Weinreich, 1953; and its theoretical and empirical development, see Boltokova, 2009).In spite of a sizeable collection of LMLS, RLS and EV literature produced in the past half century, there is evidently no magic formula to assure a language will be maintained and to allow predictions of concrete instances of LS.
It is unfortunate but true that Bangkok is a place in which LS toward Standard Thai and the efforts to maintain HLs are not seen as pressing issues (Own fieldwork data, 2010(Own fieldwork data, -2013)).Monolingual or otherwise bi-/multilingual Thais of Chinese descent in Bangkok express a shift in ethnic identity from Chinese, ethnic Chinese and Sino-Siamese/Chinese-Thai to Thai concomitant with shifting from HLs to Standard Thai.
Following the recent trend that challenges and questions conventional social categories and sociological variables that affect LS, a number of theories and key concepts inform this study, notably, the concept of "ethnic identities."The researcher agrees with Koning and Verver (2012) that ethnicities can be mediated, negotiated and reproduced by means of sociological means, discourses and practices of nation states.As such, ethnic identities are by no means stable.Drawing on this construct of ethnicities, the researcher reinterprets the sociolinguistic literature pertaining to ethnicities.There is a substantial body of sociolinguistic literature on the intimate link between language and ethnicity, ethnic identity and ethnic relations (general views of this line of research can be found in Gatbonton et al., 2011;Hoffman & Walker, 2010;Jeon, 2010).There are a number of explanatory hypotheses proposed for the relationship between ethnic orientation and linguistic variation, which may bear on an individual's or a speech group's acquisition and learning of L1s, HLs, second and foreign languages.
The majority of previous studies have largely focused on influences of ethnicity, ethnic identity, and ethnic group affiliation on LMLS of minority-language groups.Numerous studies have examined ethnicity and other factors related to linguistic variation and change.In the case at hand, it is widely recognized that Bun's and Kiong's (1993) is a major account of the process whereby the Chinese are assimilated into Thai society and their adoption of Thai ethnicity.(Another illustration of the ethnic identity of Sino-Siamese is seen in Tong's & Chan's, 2001, highly regarded book).To see the whole picture, there is a need for research exploring the intersection between generational changes in ethnic-linguistic identities and LS.The present study may help do justice to this issue.
A second approach that guides this study stems from the interdisciplinary field of social inclusion/exclusion.The role social inclusion/exclusion plays in LMLS is a relatively new research area.It is argued that the language ideology of monolingualism as enforcing the use of an official state language mediates social inclusion/exclusion in ethnically and linguistically heterogeneous nation states (For a fuller discussion of linguistic diversity, language ideology and social inclusion, see Piller & Takahashi, 2011).A fundamental question of this study is what impact has discursive practices of social inclusion/exclusion had upon the vitality and the LS of HLs? Reporting on the basis of a large-scale project in the communities studied, the present article provides a selective view of theoretical issues and empirical findings relating to the question of social inclusion/exclusion and LS.
In addition to factors such as the generational change of ethnic-linguistic identity and the social inclusion/exclusion effect, more breadth and depth to the discussion of facts and theoretical issues concerning the groups in consideration are offered in the results and discussion section (section 5).Merriam's (2009) qualitative research and Boeije's (2010) grounded theory methods provide the methodological framework for the study.

Research Questions
In this contribution, this study seeks to answer the following questions: Research Question 1: What are the language-usage patterns amongst the communities under study?
Research Question 2: What are the present perceptions of the vitality and sustainability statuses of Chinese variants-HLs-amongst the groups discussed?
Research Question 3: What appear to be the sociological factors undergirding the present status/vitality of HLs and contribute to LMLS?

Participants
The criteria for the inclusion of the sample are as follows: Au fond, the problematic ethnic identities of the sample lies in the fact that both Chinese and Thais are not themselves fully homogeneous (For a sociolinguistic view of language diversity in Thailand, see Premsrirat, 2007).The sample was recruited on the basis of linguistically heterogeneous dialect groups (8 speech groups identified).Moreover, in as much as the newer Chinese migrants (arrived in Thailand in 1970(arrived in Thailand in 's-2000's) 's) have not received the attention needed, the researcher is particularly interested in surveying these newly settled FDIs.Further, the geographical aspect of the ethnic Chinese in Bangkok and those in provincial Thailand are demonstrably dissimilar in numerous respects, including their varying degrees of assimilation into Thai society and varying degrees of proficiency in their HLs, Mandarin and Standard Thai.These factors largely involve the nature of interactions with Thais considered under the rubrics of all social transactions, convergences and divergences in local, political, social and economic communities.However, the researcher is primarily interested in the present vitality of Chinese variants and sociological effects in LMLS in the Bangkok communities under discussion.
Representing 8 Chinese dialect groups, 320 participants agreed to participate in this study in view of having given express and informed consent.On the basis of self-classification, there were divided into three ethnic-linguistic categories: Thais of Chinese descent, ethnic Chinese and overseas Chinese.Nonetheless, the sample (N=320) represents approximately less than 1% of the total target population (3.5-8 million).All ethnic Chinese participants were descended from migrants from Mainland China more than 200 years ago (approximately 4-5 generations) and their families have been living for 2-3 generations, except for those in the Taiwanese, Hong Kong and Singaporean portion of sample (first-to-third generation migration in 1970's-2000's).By the same token, participants recruited from centuries-long generational ethnic Chinese communities, are relatively demographically similar, except for the sample of FDIs (group 5 and 6 in Table 4).

Primary data
Secondary data 8 Chinese dialect groups (N=320 participants) in Bangkok were researched on the basis of a questionnaire survey and were observed and interviewed over a 3-year period.

Instruments
The three instruments adopted were: 1) a questionnaire survey; 2) an observation protocol; and 3) an interview protocol.The target populations in the different districts of the survey site were researched using the questionnaire survey, observations (participant and non-participant) and interviews (semi-structured).A 55-item language usage and subjective EV (age, domain-and media-based) questionnaire survey was administered to participants, thereby allowing the drawing of a macro-picture of language-use and LMLS patterns.The survey instrument is by no means a standard questionnaire, but is particularly designed such that it is consonant with the research purpose and suitable for the participants.The questionnaire survey consisted of four sections.The first section requests demographic information (age, educational attainment and socio-economic status).In addition, the second-fourth sections involved a sociolinguistic survey whose purpose is to allow measuring self-reported language-use patterns and self-rating language-proficiency levels.The survey questions elicit information concerning language variants commonly used in addressing interlocutors from intra-ethnic-and inter-ethnic interactions across communicative domains (The communicative domains include family/home, friendship, mass media, religion, school/education and transaction/business, as derived from Platt and Weber, 1980).Finally, the varying degree of language proficiency claimed by participants is surveyed through the asking of pertinent questions.Because self-reported data of language use and language proficiency may be inadequate indicators of actual language behaviors, explanations and interpretations derived from data should be read with due caution.Nevertheless, self-reported data are supported by observational data.

Procedures of Data Collection and Analysis
Multiple methods and three rounds of data collection were employed: 1) Round 1: Questionnaire survey; 2) Round 2: Observation; and 3) Round 3: Final retrospective interviews (1-on-1 and/or focus group) Data collection was primarily conducted in Mandarin, various Chinese dialects and/or English by the researcher with the assistance of a team of Thai-English bilingual translators.Participants (200 out of the total number of 320) completed the first two rounds of data collection (survey questionnaire and observation), with only 75 of them participating in the final (3rd round) retrospective interviews.The first round of questionnaire surveying involved the participants' experiences of daily language-usage patterns, vitality perceptions/sustainability statuses of HLs (Chinese dialects) and reasons for LS toward a habitual use of Thai and/or Mandarin.The subsequent two rounds of observation and interview explored more in-depth topics and issues relating to the responses gathered from the first round of questionnaire surveying.The interviews, using a pre-agreed interview protocol that outlines questions to be asked, were successful in fostering further opportunities for researcher-participant interaction and gathering insightful reflections by the participants.To encourage more elaborate and personalized responses while surveying and interviewing the participants, the language used by the researcher and assistants/translators was not scripted but semi-structured.As it turns out, a total of 280 questionnaire surveys were returned to the researcher (return rate of 87%).A total of 150 observational field notes were gathered during the field visits.A total of 130 audio-recorded interviews were conducted, yielding approximately 60 hours of material.
Data derived from questionnaire responses, observational-field notes and the digital recordings of interviews were coded, transcribed and triangulated for analysis.While analyzing the data, emic and etic approaches were adopted to enable the researcher to balance the participants' and the researcher's views.Data were analyzed with a careful reading allowing for the classification of recurring themes (and recurring sub-themes), utilizing the following five categories: (1) Research purposes (see sub-section 1.2); (2) Research questions (see sub-section 4.1); (3) EGIDS, a reformulation of Fishman's (1991) 8-level scale, termed Graded Intergeneration Disruption Scale (GIDS); a 13-level scale, termed Expanded GIDS (as reformulated by Lewis and Simons, 2010) in Table 6; (4) the conceptual and methodological underpinnings for the generational change involving ethnic-linguistic identity theory and social inclusion/exclusion theory; and (5) a statistical tabulation of numerical data derived from responses to the questionnaire survey.In addition, the frequency and the percentage of the occurrences of HLs (Chinese variants) were analyzed by means of summing up their total presence in 46 possible scenarios of daily language uses across communicative domains: family/home (8), friendship ( 14), mass media (10), religion (4), school/education (4) and transactions/buying and selling (6) (The communicative domains are based on the study conducted by Platt and Weber, 1980; this round of data analysis procedures is a replication of Ting's and Ling's study, 2013).Sources: Adapted from Fishman, 1991;Lewis and Simons, 2010 Note.GIDS corresponds to EGIDS (EGIDS is an expanded and reformulated version of GIDS)

Question 1 (Language-usage Patterns of Intra-and Inter-ethnic Communications)
The results show that the official state language-Standard Thai-is the most prevalent, because it is the most frequently used code (more than 2,000 reports), followed by the regional hegemonic language-Mandarin (approximately 700 reports).The use of the respective HLs (Chinese variants) is higher in intra-ethnic communications (approximately 800 reports) than inter-ethnic interactions (approximately 50 reports).The results also show that various Chinese dialect groups fall into four categories.The first category of passive bi-and multi-linguals comprises individuals and communities of first-and second-generation overseas Chinese/FDIs, using their respective HLs (Chinese variants) and/or the regional language/Mandarin more frequently than other languages across domains.Participants in this category show their overt preferences and strong ideological commitment toward the use of their respective HLs and/or Mandarin.In spite of some in this category who are classified as Chinese-Thai passive bilinguals (far more extensive use of HLs than Thai), their daily language-usage patterns demonstrate that they are nearly Chinese monolinguals.Participants in the category of passive bilinguals include first-and second-generation overseas Chinese/FDIs from Taiwan and Hong Kong, excluding Singapore.
The second category of active bi-and multi-linguals includes first-and second-generation Yunnanese or Yunanese-Thais, second-and third-generation ethnic Chinese/FDIs from Hong Kong, Singapore and Taiwan, and third-fifth generation Thais of Chinese descent (Cantonese, Hainan and Teochews).Some participants of third-fifth generation Thais with Chinese ethnic roots in this category received education (secondary or university-undergrad and/or grad) in Mandarin or acquired Mandarin during their upbringing in a Chinese-speaking state (PRC or ROC).Others of third-fifth generation Thais with Chinese ethnic roots (Hainan and Teochew) in this category received higher education (university-graduates) in English-speaking nation states (e.g., Australia, U.S.A., and U.K.).Most participants in this category with a mean age of late forties use a global language (English) or a regional hegemonic language (Mandarin) parallel with the official state language-Standard Thai-in work domains.However, the younger third-fifth generation Thais of Chinese descent (Hainan and Teochew) or part Chinese in their early twenties learn English in Thailand.For the ethnic Chinese entrepreneurs in the business networks whose ethnic languages are not Teochew, it is striking to see that they tend to use Teochew more frequently than their non-entrepreneur counterparts who are of Teochew descent.For the second-generation Singaporean-Chinese sample/FDIs, they are in the category of active bilingualism, in as much as they fluently speak Singaporean English (Singlish) and have become accustomed to using Standard Thai.
The third category of Thai monolinguals includes third-fifth or later generation Thais with Chinese ethnic roots who have lower levels of educational attainment and have a much lower socio-economic status.Both qualitative and quantitative data reveal that three main trends were discernible in the answers to research question 1.The communities in question, as a whole, have been shifting away from HLs (Chinese variants) toward Standard Thai.In addition, some community members under consideration have been shifting from HLs toward Mandarin.Further, there is an increase in the use of English by members of the younger age group to the middle-age members of the communities.The second and third trends were ill-defined and, more accurately, were under addressed by previous studies (e.g., Morita, 2003Morita, , 2007)).Given the status of English as a global lingua franca and the role of Mandarin as a regional hegemonic language (with a base in Asia), it should not be surprising to see the second and third trends of a gradual LS away from HLs.

Question 2 (Vitality Perceptions/Sustainability Statuses)
The quantitative analysis, presented in Tables 7-10 below, shows that the vitality perceptions/sustainability statuses of Cantonese, Taiwanese and Yunnanese are higher than other groups.The qualitative data also report that Taiwanese was used at home, friendship and religions domains (for example, churches).Yunnanese, nonetheless, was mostly used at home and in intra-ethnic communication with relatives and business partners.Cantonese was used solely at home for 45-64 and 65 + age groups, but not for other age groups.Applicable to all Hainan (some) and Teochew (some)

Applicable to few
Table 11 presents spoken and written modalities used by participants across communicative domains.It is essential to note that some in 65 + and 45-64 age groups, particularly those who are first-generation immigrants from various dialect groups, read Mandarin newspapers and watch Mandarin channels.In sharp contrast, 2-14 and 15-29 age groups are more likely treat Mandarin as a school subject in the Chinese heritage school.They most likely engage in mass media outside the school in Thai.
Home is the first domain to be considered in Table 11.Joshua Fishman is largely responsible for the home domain theory.In his view, the home is crucial for ethno linguistic minority speakers to ensure the intergenerational mother-tongue transmission (Fishman, 1991(Fishman, , 2001)).He has argued for two to three decades that minority-language speakers cannot bypass their home if they are to maintain their HL.In spite of the strong contrary arguments by Fishman and others, Teochew participants showed that they bypassed their home and yet maintained Teochew proficiency.(Thailand-born second-generation Teochews or part Teochews did not even talk to their first-generation Teochew immigrant parents in HL).However, a decade ago, Thailand-born second-fifth generation ethnic Teochew learned their LM mainly in business domains.Nonetheless, today Teochew does not function as the lingua franca in intra-ethnic Chinese business communication.Thai has gradually replaced Teochew in the intra-ethnic business networks of Chinatown (Yaowarat).Table 12 presents the stages of language endangerment of the various Chinese dialect groups studied.One major obstacle to maintenance for Cantonese (estimated stage 7-9), Hainan (estimated stage 9), Hakka (estimated stage 9), Hokkien (estimated stage 9) and Teochew (estimated stage 8) groups is exposure.They cannot maintain their respective HLs, in as much as they did not have enough exposure across communicative domains.

Question 3 (Reasons for Maintenance and Shift)
In addition to factors identified by previous scholars (e.g., Morita, 2003Morita, , 2007) ) that caused LS from HLs to Thai, this study has found new reasons for this phenomenon.

Extensive Ethnic Network, Multinational Business Industry, Institutional Support, and Bilingualism and Diglossic Code-mixing
In contradiction to Skinner's assimilation theory (1957,1958,1963,1973), some salient examples of third-fifth (or later) generation ethnic Taiwanese, Singaporean-and Yunanese (and fewer Teochews and Hainanese) studied showed signs of bilingualism and diglossic use of codes across communicative domains.Most of the participants who showed no obvious signs of LS from HLs to Thai and/or from Mandarin to Thai were constantly in contact with extended family members and business circles in their respective homelands.By the same token, they were characterized as transnational, bi-and multilingual merchant families, in as much as their code uses were conditioned by the increasingly extended ethnic networks and regionalized business industries, and institutional support from PRC or ROC, and social models that accompanied them and were re-created by them.
Some ethnic Taiwanese were salient examples.They grew up attending Chinese heritage schools in Bangkok run by ROC and returned to Taiwan to receive undergraduate education.They were dating prospective marriage mates in Taipei.Some ethnic Teochews received both undergraduate and postgraduate education in ROC or PRC and devoted their careers to teaching Mandarin in Thai institutions of higher education.A smaller number of ethnic Hainan pursued undergraduate programs in PRC and returned to Bangkok and served in temples devoted to the shrine worship of ancestors.Some ethnic Yunanese received strong institutional support from ROC and moved back and forth between Northern Thailand (where they received institutional support) and Bangkok.Other ethnic Yunanese were involved in Mandarin-Yunanese bilingual business circles as full-time traders with the PRC.In recent decades, some Yunanese Thai spent more months in the PRC and fewer months in Bangkok in managing their multinational business empire.There was increasingly more and more Taiwanese Thai and Singaporean Thai who took turns in living in Taipei and Bangkok, and Singapore and Bangkok while managing trans-Pacific trade.In summary, these examples do not support the Skinnerian paradigm of complete assimilation of Chinese in Thailand after the third generation.
However, there are several issues in this sub-section's current form.Due to space limitations, the author has not provided sufficient context based on the literature to support claims concerning the types of extensive ethnic networks, multinational business industries involving trade with the PRC and institutional support from the ROC that enhance the LM of HLs and Mandarin.For a review of Yunanese in Thailand and their institutional support from the ROC, see Huang (2005).For a discussion of Yunanese Chinese trade in Thailand, see Hill (1998).It is widely recognized that second-fifth (or later) generations of ethnic Chinese or those of part Chinese who identify themselves as Thais of Chinese descent or claim to be 'pure' Thais.Such a generational change in ethnic-linguistic identity (it only takes one generation) plays a role in the extent of LS from HLs to Thai.One may consider that if I speak Thai as L1, I am a Thai.This view uses the level of linguistic proficiency in determining one's ethnic-linguistic identity.However, recently gathered data (2010-2013) epitomize the view that 'because I am a Thai, I speak Thai,' in spite of the fact that a small number of second-generation Chinese Thai were found to be late bilinguals (in as much as their Mandarin was better than their Thai).In this view, ethnic identity foregrounds linguistic proficiency for some second-fifth (or later) generation ethnic Chinese (not applicable to all).For a review of 'flexible identity' of Chinese-Thai communities, see Auethavornpipat (2011).
Additionally, the findings reveal some inconsistencies with previous research (e.g., Morita, 2003Morita, , 2007)).Morita (2003, p. 492) left the impression that only Chinese elites in business and political sectors who held high positions in Thailand encouraged assimilation into Thai society (LS away from HLs to Thai is seen as assimilative acts) so as to obtain greater prestige, power and wealth.Stated another way, Morita (2003) may unconsciously fail to take into account those who did not belong to elite groups.
This study is devoted to members of the middle-and lower-classes, as well as members of higher-classes, by examining their expressions of ethnicity and LMLS.Findings in part agree with Morita (2003) that higher-class Thais of Chinese descent has embraced the Thai ethnic-linguistic identity.However, data also show that numerous middle-and lower-class participants have undergone or are undergoing changes in their ethnic-linguistic identity in relation to their LS, except for first-generation FDIs and academics from Taiwan, Hong Kong and Singapore, who view themselves as overseas Chinese, not Thais.
The researcher found that some middle-and lower-class participants express the non-relevance or denial of traditional Chinese cultural ideologies and/or rejection of Chinese ways of doing business and linking this to their LS.Even participants in family-run businesses or ethnic Chinese entrepreneurs disliked some inherited cultural values.'The Chinese way of doing business is to invest in everything of anything that you think you can turn a profit (even if you don't like the industry you are in), work very hard, don't have many holidays or even don't have time to rest, in order to save a lot of money to buy a big Mercedes-Benz.'This quote is derived from one second-generation ethnic Chinese participant and it reveals a generational change in beliefs and values of ethnic Chinese entrepreneurs.In agreement with Koning & Verver (2012), this study proposes that there has been a generational change in ethnic-linguistic identity occurring amongst the communities researched.The younger age groups may or may not appreciate hard work and do not necessarily view earning money as the primary goal of life.Simply put, they do not want to be like their parents or grandparents in their business conduct.Such a generational change of ethnic-linguistic identity, along with the LS, does not create a problem in practicing traditional Chinese family festivals and rituals, in as much as they use Thai in such cases without using HLs.
Some scholars (Morita, 2003, p. 492) have claimed that Mainland Chinese do not accept Thailand-born Chinese.This allegedly generated LS away from HLs and the embracing of a Thai ethnic identity.To bring this topic current, the data gathered for the present study show that Mainland Chinese in recent years (2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)) not only accept Thailand-born Chinese but also desire to become Thai permanent residents or citizens, because many Mainland Chinese desire to migrate to more developed states and Thailand is one of the choices.
Also, the earlier Chinese diaspora to Thailand (commencing in the first half of twentieth century) is characterized as embodying a classical push-pull migratory model in such that they were most likely to adopt a new ethnic identity as Thais in order to enjoy abundance in Thailand in view of widespread famine in their homeland.In contradiction to this view is the view that first-generation Chinese immigrants intended to return to their homeland with the more current evidence presenting challenges to the widely-accepted concept of 'overseas Chinese.' 5.3.3Discursive Practice of Social Inclusion/Exclusion and Language Shift 'If my daughter does not and cannot speak Thai, she will be in big trouble (being discriminated against) in schools and later in the job market.To be socially included, one must speak Thai.Thais don't like foreigners who [are not tourists] who do not know Thai ways and do not speak basic or conversational Thai.'This quotation is derived from one first-generation Chinese participant studied and reveals a widely accepted belief of community members.
Thailand was never colonized by Eastern or Western states in contrast to its neighbors and today it represents a strong economic giant in mainland Southeast Asia-a desirable market for FDI (For a review of Thailand's business attraction and economic growth, see Niffenegger et al., 2006).For the aforementioned reasons and others, pride in being Thai has become wide spread.As such, a strong Thai in-group identity is a matter of concern for first-generation Chinese immigrants, as well as second-fifth (or later) generation Thais with Chinese ethnic roots.Speaking Thai is an in-group marker, thereby the massive majority of participants preferred their children to speak Thai as their L1 and to behave in the Thai way.It must be added that when there is conflicting cultural imperatives between Thais and people from other cultures, the former prevails in Thailand.Thai supremacy was the mechanism that operated in job negotiations with high-skilled western academic migrants (witnessed by the author).'If there is a way to do things in Thailand, the Thai way is the only right way.Thais are less concerned about your (non-Thai) ways.They only do things in their way, regardless of how in-efficient it is.'This quotation is derived from one first-generation Chinese participant and he noted that this widely accepted belief in Thai privilege plays a major role in the extent of LS away from HLs to Thai for the communities under study.
Additionally, Chinese-accented Thai was stigmatized by the first-second generation ethnic-Chinese participants researched.Therefore, numerous earlier generations of ethnic Chinese attempted to ensure that their children speak Thai without phonological interference from Chinese pronunciation.This is the primary reason why they were hesitant to transmit HLs.Surprisingly, this major reason for LS was not explored by previous scholars (e.g., Morita, 2003Morita, , 2007)).
Furthermore, speaking the Thai language involved more than the need to be socially included for first-generation Chinese immigrants.It was adopted as a disguise strategy-a means of survival by which illegal Chinese-speaking immigrants pretended to be Thai-speaking Thais, in an attempt to avoid detention and deportation.

Religious Tolerance, Intermarriage and Language Shift
Thailand, in comparison with Indonesia and Malaysia, is more religiously tolerant and less hostile toward the Chinese diaspora, notwithstanding their greater or lesser resentment against ethnic-Chinese economic dominance.Over the centuries, Thais have welcomed and are continuously welcoming numerous foreign gods (for example, Hindi elephant gods) and religious missionaries (for example, American and Korean protestant pastors).The immigration of Chinese Taoists and their religious practice of worshiping at ancestral shrines were presumably facing more resistance in Muslim-dominant Indonesia and Malaysia than by Buddhism as practiced in Thailand.Ethnic Chinese in Thailand do not need to live their lives in accordance with Islamic beliefs.Thais widely accept ethnic Chinese family tradition such as the tomb-cleaning festival and autumn/moon-cake festival.Viewed from a socio-historical perspective, some researchers (e.g., Guoto, 2012) also pointed out the 'cultural similarity' between China and Thailand as an enhancing factor in Sino-Thai relations.In summary, religious tolerance in Thailand and similar cultural practices shared by China and Thailand shed light on the reason why the Chinese are accepted culturally in Thailand and their religious practices are relatively easily accepted in Thai society.
(Also, notice that some Thais are now becoming adherents of Chinese Mahayana beliefs.)Hence, Thailand as demonstrably becomes a new homeland to the best-integrated overseas Chinese community in Mainland Southeast Asia.Shockingly, these cultural-and religious-similarity factors were under explored by previous researchers (e.g., Morita, 2003Morita, , 2007) ) who have addressed the extent of language shift of ethnic Chinese in Thailand.
In view of the aforementioned and other key factors, more overseas Chinese and ethnic Chinese have intermarried with Thais in Thailand than in Indonesia and Malaysia.A reason behind the language choice of Chinese-Thai couples is convenience.The medium of communication between a Thai-speaking Chinese husband and a Thai-speaking Thai wife is carried out in Thai, because it is more convenient to communicate in this code.The children and grandchildren of these Chinese-Thai couples grew up using Thai as their L1.Although Hainan and Hokkien are used for father-mother (husband-wife) communication in some intra-ethnic marriages, Thai is used for parent-child cross-generational communication.For a review of intermarriage between Chinese and Thai in the early twentieth century, see Morita (2003, pp. 486-487).

Heritage Language Education and Language Shift
Some previous research (e.g., Morita, 2003Morita, , 2007) ) was unclear about what languages (whether Chinese variants or Mandarin) were actually taught in Chinese heritage schools in Thailand.Furthermore, some scholars (e.g., Morita, 2003Morita, , 2007) ) mistakenly implied that Chinese heritage schools played a role in the LM of HLs (Chinese variants) for second-fifth (or later) generation ethnic Chinese.Thus, their research showed that the decline of Chinese heritage schools were partially responsible for the LS from HLs toward Thai.The current study, nonetheless, shows otherwise.Chinese heritage schools have been increasingly promoting the LS from HLs toward Mandarin, instead of the LM of HLs.It must be acknowledged that possibly some Chinese heritage schools aided in the LM of HLs more than half a century ago, but they have not done so in the last twenty-thirty years.Therefore, it is likely that Chinese heritage schools play a major role in shifting the communities' younger age groups from HLs toward Mandarin.For a discussion of the effects of the Mandarin-medium in Chinese heritage schools on students in Bangkok, see Morita (2007, p. 55).

Exposure, Mode of Settlement and Language Shift
As noted earlier, exposure plays a key role in LS away from HLs toward Thai, because there were apparently shrinking domains of the use of HLs amongst the communities discussed.The exposure to the Thai code instead of the Chinese code has to do with mode of settlement.Commenced from the middle of the 19th century, commercial districts in Bangkok were inhabited by large numbers of eastern-coastal Teochew-and-southwestern Yunanese-Chinese immigrants (Bruthiaux, 2008, p. 138).Even though there was a concentration in the mode of settlement, the code anticipation (or the code demand) in these business districts resulted in the use of Thai as a convenient lingua franca for both intra-and inter-ethnic business transactions.

English and Language Shift
One key factor in combination with the aforementioned reasons that contextually led the communities discussed into LS away from HLs but not necessarily toward the direction of Thai is the changing focus on English amongst the middle-and younger age groups researched.The massive majority of participants (excluding the senior age group) learned English as their first foreign language.Many of the middle-age participants revealed that when they were young, they stressed English at the expense of Mandarin and HLs.The same dynamic is at work nowadays vis-à-vis the sample of children, adolescents and young adults studied.Unsurprisingly, they all expressed their interest in English instead of their HLs, particularly those English-educated ethnic Chinese academics in Thai higher education.It is apparent to even a casual observer who pays a visit to Bangkok's Siam Square to witness numerous English tutoring centers and an increasingly growing number of English-medium programs, international schools and international colleges.The focus on English may or may not directly lead to a LS toward English.Nonetheless, it has an enormous effect on the shift of the focus on the selected code to be added to one's repertoire.

Summary and Concluding Comments
In conclusion, this article addresses an important set of issues that were insufficiently researched in previous studies-the language use and vitality of ethnic Chinese societies in the new diaspora-recent Chinese immigrants entering Thailand in 1970's-2000's.Additionally, in this article, the researcher examines the discrepant levels of HL proficiencies across Chinese dialect groups.Furthermore, it contributes by making more current the sociolinguistic literature on Sino-Siamese societies by providing new explanations for LMLS even if the factors newly uncovered by the researcher was in existence long ago but were unidentified by previous scholars in this area.
Some limitations of the present study have to be acknowledged.This paper does not sufficiently contextualize reading and writing behaviors of the communities under consideration if only because it mostly focuses on speaking behaviors.
In addition, generational LS from Chinese-Thai bilingualism to a habitual (but not exclusive) use of Standard Thai has been occurring in the second-generation of ethnic Chinese and the third-fifth (or later) generations of Thais of Chinese descent as a whole, whilst such a shift has not always occurred in first-generation overseas Chinese.Additionally, LS away from HLs (Chinese variants) toward a habitual use (not exclusive) of Mandarin has also been occurring in some members of the first-fifth (or later) generation participants discussed, but not all.This shift to Mandarin has been insufficiently addressed extent by previous scholars.Amongst all Chinese dialect groups examined, they all underwent various degrees of LS to Thai, particularly the Teochew group as evidenced by a relatively more serious level of discontinuation in intergenerational transmission than the relatively well-maintained Cantonese, Taiwanese and Yunanese.This discovery challenges the traditional belief that Teochew's vitality has been strong, in as much as its speakers at present still outnumber all the other ethnic Chinese combined.
Some of the most prominent themes derived from findings are listed below.One of the many new findings generated is that ethnic Chinese's LS away from HLs to Thai can also be interpreted as resistance against the traditional ideology of ethnic entrepreneurship-the Chinese way of doing business-that positions them to work hard for money as their life-long goal without taking into consideration their job interests.
Another atypical finding is that some participants studied bypassed home by acquiring Teochew in business domains, because parent-child communication was carried out by Thai-speaking Siamese mothers.This particular finding questions Fishman's classic home domain theory (1991,2001).
One particular salient point of research results is as follows.In most cases, Hokkien and Teochew were acquired by younger age groups and not from parent-child communication as the classical intergenerational transmission theory would suggest.Rather, they acquired their respective HLs through grandparent-grandchild communication, because they were raised by their HL-speaking grandparents instead of their Thai-speaking parents.For lack of better terminology, such an atypical practice (nearly unseen by western scholars) is termed 'jumping-,' 'skipping,' or 'passing-one-generation' transmission of HLs.
Another salient finding is associated with the increasingly regionalized business network in the Asian-Pacific region.At present, some first-to-third generation ethnic Taiwanese, Singaporean and Yunanese divide their time in two nation states, thereby living in two countries as their homes.Because of their strong ties to the FDI network, they are native to both the country of origin and the state in which they make investments.This finding questions the validity of William Skinner's (1957Skinner's ( , 1958Skinner's ( , 1963Skinner's ( , 1973Skinner's ( , 1996) ) fully assimilative theory (Recent scholars, such as Chan Kwok Bun, Tong Chee Kiong, 2001, and Morita Liang, 2003, 2007, also argue against William Skinner).
As for the future, HLs (Chinese variants) will be unlikely to be reversed for the Chinese dialect groups discussed which have undergone LS many generations ago.However, for viable groups who have maintained their HLs for decades or centuries (a small number of middle-and older-aged Cantonese, and the majority of ethnic Taiwanese and Yunanese), their Thai-speaking wives and in-laws, as well as Thai television channels, are the challenges faced in their LM of HLs for their children in the home domain.
More interestingly, there is increasingly more and more middle-and younger-aged ethnic Chinese interested in learning Mandarin to be used in business domains (instead of a language used at home).For the business-motivated reason rather than the heritage-motivated reason, Mandarin more likely has a brighter future than HLs of the communities researched.This view is in agreement with Morita (2007, p. 55).
Overall, the topics introduced comprise a relatively new research area, involving the filling in of gaps in the sociolinguistic literature on Sino-Siamese communities.We are yet to know the future of stronger HLs and Mandarin in the communities discussed.However, it is certain that amongst second-to-third (or later) generation ethnic Chinese as in the case of the FDIs researched, they are characterized as bi-and multilingual and increasingly undergoing struggles to redefine and renegotiate their ethnic-linguistic roles within the rapidly changing national capital of Thailand.

5. 3
.2 Generational Change in Ethnic-Linguistic Identity and Language Shift 'I think I am an [overseas] Chinese, definitely not a Thai.After I retire from my work in Thailand, I want to go back to [retire and die in] Taiwan-my home country.However, after my Thailand-born daughter grows up, she may want to stay in Thailand permanently with her [Thai] mom because she thinks she is Thai.'This quotation is from one first-generation Taiwanese participant studied and it reveals the generational discrepancy of perceived ethnic identity between a Taiwanese [Chinese] father (a self-identified overseas Chinese) and his Siamese[Thai]   daughter (a self-proclaimed Thai of Chinese ethnic roots).

Table 1 .
claims to be of Chinese descent (CIA-The World Percentages in total Sino-Siamese population

Table 2 .
Demographic data for Chinese entering Thailand

Table 3 .
The diversity of Chinese dialect groups in Bangkok Sources: Own fieldwork data,

Table 5 .
Profiles of participants

Table 7 .
Summary of general language-usage patterns (Approximate frequencies/reports of language use across communicative domains)

Table 9 .
Summary of self-rating proficiency index of HLsData in Tables 7-10, suggest that Hainan, Hakka, Hokkien, Singaporean Southern-Min and Teochew have significantly lower vitality perceptions/sustainability statuses (closer to extinction).Even though the fact that Hakka has a relatively stronger institutional support and some revitalization efforts were carried out in recent years(2005)(2006)(2007)(2008)(2009)(2010), Hakka has been widely lost due largely to the discontinuation of intergenerational transmission.Hokkien and Teochew were unsurprisingly spoken by a handful of 15-29 age groups, in as much as they were raised by Hokkien-and Teochew-speaking grandparents.It should be noted that the use of the Southern-Min Singapore variety had undergone a shift toward the Singaporean English (Singlish) prior to the resettlement of speakers in Thailand.

Table 10 .
Measurement of language vitality by age groups

Table 10
presents the overall patterns of language use as predictive of language vitality by age groups.It must be acknowledged that across all age groups, all acquired native proficiency in Thai except for some overseas Chinese-first generation FDIs from Taiwan, Hong Kong and Singapore.Amongst those who speak Thai as their L1, the older age group (65+) speaks fluent HLs and Mandarin, but speaks almost no English.The middle-age group (45-64) has a great divide.Some are highly proficient in Mandarin, as well as in HLs (Chinese variants), whereas others speak very limited or no Mandarin and HLs.The younger age group (2-14 and 15-29) is more likely to want to learn Mandarin and English without taking HLs into consideration.

Table 11 .
An evaluation of predominant medium of communication

Table 12 .
A general summary of vitality perceptions/sustainability statuses