論英文社會形象及其意識形態：與英文所有權之關聯

全文

(1)國立臺灣師範大學英語系碩士論文 Master Thesis National Taiwan Normal University Department of English. 論英文社會形象及其意識形態：與英文所有權之關聯. Language Ideology of English: Its Relation with Linguistic Ownership. 指導教授：蘇席瑤 Advisor: Dr. Hsi-Yao Su 研究生：李婉歆 Student: Ann Wan-Hsin Lee. 中華民國 101 年 7 月 July, 2012.

(2)

(3) 摘要本研究旨在探究全球化影響之下英文在臺的社會形象(social images)、觀感 (perceptions)、以及臺灣民眾的英文意識形態(language ideology of English)，由過去文獻已知英文被賦予十分正面的社會價值(e.g. Chen 2006; Lee 2008; Tsai 2010; Wang 2000)。然而，過去研究鮮少探討臺灣民眾對英文的意識形態，或以英文能力為變因，討論不同的英文觀感和意識形態。藉由訪談，本研究討論英文的意識形態如何因為英文所有權，即能力與否，而有顯著差異。本研究共有 38 位受訪者，其中，有 14 位英文使用者(English users)和 24 位非英文使用者(non-English users)。訪談的總時數是 24 小時 18 分鐘。鑒於近期的研究顯示，訪談內所透露出的互動和訊息不應被視為不自然(artificial)和線型 (linear) (e.g. De Fina 2011; Worthem et al. 2005)，本研究把訪談語料依其結構和相關性分成兩大類。第一類為受訪者對訪談問題的直接回應(direct responses)，第二類則是由訪談問題所導出受訪者相關的過往經驗、反映及想法(semi-direct responses)。相較於直接回應，部份相關的回應有較高的可靠性，因為受訪者對於對話內容以及訪談流程的干預明顯降低。研究結果指出，在訪談的直接問答之間，所有受訪者皆強調英文的必要性，此主流意識形態可從受訪者對學習英文的肯定及對於英文能力不足者的貶抑 (deprecation)窺知。此初步的研究結果大致呼應已知的文獻，代表英文以及它被賦予的正面形象已是廣為認同、接受、並視為理所當然的社會價值體系裡一環。除此之外，受訪者所透露出對於英文在就業上的幫助也顯示英文在臺灣社會有著高工具性(instrumental)，並在個人職業生涯扮演舉足輕重的角色。這象徵英文是一種資本(capital)。由於英文被視為職場和學術利器的形象鮮明，雖英文使用者和非英文使用者看似對英文抱有相同概念(conceptulization)，英文使用者和非使用者無可厚非地在全球化的社會之下定位不同。相較之下，間接回應顯示英文的接受度在非正式或同儕對話中反而明顯降低，且個人的英文能力和英文的所有權並非自稱。英文使用者和非英文使用者都會以主觀的條件審視英文及說話者進而認證或否定英文的所有權。此外，雖英文被視為必要，但是，對話的情境決定語言的選擇(code choice)，不適當地顯露英文能力反而容易招致負面觀感。研究結果也顯示英文使用者和飛英文使用者都會為了語用的需求在日常對話中穿差英文，但有趣的是受訪者皆一致認定英文使用者在非正式場合下有責避免使用英文。研究結果發現，英文能力與否影響說話者在言談中的定位，同時也影響英文的意識形態。本研究以英文能力為變因探導英文的社會形象和意識形態，由於時間和區域限制和等因素，未能收集多方語料以更深度探究本文的研究目標。在未來研究中若能補足語料收集上的不足，將會對此研究方向及結果有所助益。. i.

(4) 關鍵字：意識形態、語言態度、全球化. ii.

(5) Abstract The study investigates language ideology of English through a qualitative analysis of interview data. It has been well-established that English is highly valued owing to its instrumental values (Chen 2006; Lee 2008; Tsai 2010; Wang 2000). Nonetheless, seldom did past studies address the ideological aspects and the possibility of English competence as a variable in shaping different language ideology. The study is aimed at looking into how English is conceptualized. The study has collected data through interviews. Among all the 38 informants who participated in the study, there are 14 English users and 24 non-English users. The total length of the interviews is 24 hours and 18 minutes. In the light of recent studies in critical analyses of interview data (e.g. De Fina 2011; Worthem et al. 2005), the study categorizes data into two major categories. Direct responses are the informants’ straightforward replies to the interview questions. Semi-direct responses refer to the informants’ reflections on events and further thoughts evoked by the interview questions, but not necessarily directed at answering the questions. It is found that English is predominantly considered essential. The prevailing language ideology of conceptualizing English as a necessity is apparent in the informants’ overt affirmation of English acquisition and the deprecation of incompetence. The findings suggest that the appropriation of English may be perceived to be an established and prescriptive social norm. it could be postulated that English is linguistic capital, an accumulated labor which could transfer to interests and values (Bourdieu 1977, 1986, 1991). Though the necessity of English is reinforced by both English users and non-English users, they positioned themselves different from the other party. Though English has been overtly affirmed, the appropriation of English and the iii.

(6) concept of English ownership are context-specific and dependent on others for the recognition. Standard language ideology, the preference for an idealized linguistic form (Lippi-Green 1997), is found to be a subjective criterion held by English users and non-English users to evaluate English and its user. Consequently, self-claimed competence in English and ownership, the legitimacy of the language (Widdowson 1994; Higgins 2003; Norton 2003), could be easily challenged and further invalidated by others. Moreover, English is perceived to be context-specific, implicating that the allocation of functions of English is distinctive compared with Chinese and other indigenous languages. The display of English competence is significantly oppressed in local contexts, possibly attributable to the public market values of English. The dispreference postulates that meeting linguistic norms in local contexts maybe equally stressing to that in public contexts (Woolard 1985). Remarkably, the burden of the accommodation falls chiefly on English users. The finding indicates that competence in English is a significant variable in influencing language ideology of English as English users and non-English users position themselves differently. Analytically, semi-direct responses show that the proximity of English is context-specific. Methodologically, the examination of the semi-direct responses demonstrates the necessity of analyzing interview data critically. The study has attempted to shed some lights on language ideology of English. Yet, naturally occurring data would definitely reveal more authentic information with higher reliability in terms of English use. Additionally, the recruitment of informants was confined in terms of the regional restrictions and the imbalanced number of informants in each social category. With these limitations taken into consideration, a well-developed account would surely be yielded. Keywords: language ideology, language attitudes, globalization. iv.

(7) Acknowledgements As I am wrapping up the thesis, I have kept thinking how I would present the acknowledgement to show my full appreciation to every single person who has supported me, encouraged me, and lent me their shoulders when the pressure and anxiety of the thesis and my teaching job application went out of control. The thesis is dedicated to you all. I would like to thank my advisor, Professor Su Hsi-Yao. I don’t honestly know where I should begin. A paragraph never comes close to expressing how grateful I am. You always show me this ultimate faith in me, the faith which I don’t think I would ever have in myself. Thank you for your faith in this normally inconfident me. Thank you for teaching me how to challenge myself. With a smile, you are always willing to listen to me babbling about practically everything. Thank you for your patience. You always say that we are special to you because we are your first bunch of students. Thank you for making me feel so special to be one of your special students. During these eight years, thank you for being such a great mentor to me (though you once mentioned that you don’t think you are) in all the ways you possibly can. The influences you have to me (and to a lot more students I am sure) is lifelong. I would also like to thank Professor Chang Miao-Hsia and Professor Chan Hui-Chen. To Professor Chang Miao-Hsia, thank your for being so encouraging. It always makes my day whenever I run into you on campus. Thank you for the insightful comments you gave me. To Professor Chan, thank you for accepting me as an auditor long time ago. Thank you for agreeing to be may committee member when you were feeling unwell. Thank you for approving the thesis. It means so much to me. A special thank you to Joy Wu, one of the best teachers that every student could have. Thank you for understanding me in a way that I myself may not be able to. Thank you for the hugs you give me when I am on the verge of tears. Thank you for giving me so many chances to make me understand that you believe in me. Thank you for being always here to hear me out whenever and wherever I need someone. Thank you for being the person that the countless thank-yous are still not enough to show my appreciation. Thank you so much! To Sam, Prince of Syntax, I promised that you would have your own paragraph in my acknowledgements, so here it is. Thank your for being one of my best pals in my graduate school life. My life in the past three years would have been boring without the karmas that we evilly generated. Thank you for the stunt at my oral defense. My eyes welled up at your appearance. Thank you for bringing me so much laughter. Thank you for being a Syntax lover. It never fails to amaze me how v.

(8) miraculous it is for us to become friends in the first place. Yes, UG is powerful. But Sociolinguistics is mighty!!! Ha. I would also like to thank my friends, who are poorly targeted by me when I let out all my frustrations and anxiety. To Betty, thank you for calming me down everytime I am bawling my eyes out. You know that I am always very grateful to have a friend like you even though I am terrible at expressing it, don’t you? To Bebe, the past year is like a roller-coaster ride to both of us. I am glad that I have someone like you who understands perfectly the low spirits that constantly shadowed over me. I wish you all my best about the new lease of life that we are heading for. To my classmates, Gina, Monica, Kathryn, and a lot more people, thank you for the warm smiles you gave me whenever we ran into each other on campus. To Zhiyou, A-Jin, Tommy, Rachel, Sparq, ad Hui-Jun, thank you for being who you are and for being there for me as you always do. Last, but never the least of course, I would like to thank my family. To my Dad, thank you for the 100% of unconditional support about every decision I have made. Your little girl has now grown up, but will always remain your little girl. To my Mom, thank your for being the person to whom I could completely open up, even though you always say that I should have acted more mature. To Roger, my younger brother, thank you for your caring messages everytime we chatted. I was born lucky. To everyone who participated in the study, thank you for your generosity. To everyone who cheered for me, believed in me, and encouraged me in every possible way you could, thank you for your trust in me. I have done it!. vi.

(9) Table of Contents i iii v. CHINESE ABSTRACT ENGLISH ABSTRACT ACKNOWLEDGEMENTS TABLE OF CONTENTS. vii. CHAPTER 1. INTRODUCTION. 1. CHAPTER 2. LITERATURE REVIEW. 7. 2.1 LANGUAGE IDEOLOGY 2.1.1.Outline of language ideology 2.1.2.Debated issues of language ideology 2.1.3 Language ideology and other related terms. 2.2 THE ECONOMY OF LANGUAGE 2.3 GLOBALIZATION AND THE RISE OF ENGLISH 2.3.1 Globalization and language 2.3.2 English as a global language. 8 8 10 11 13 15 15 16. 2.4.1 English Education in Taiwan. 19 22 22 23. 2.4.2 Other Institutional Power. 25. 2.3.3 English learning, social implications, and concerns 2.3.4 Linguistic ownership. 2.4 ENGLISH IN TAIWAN. 27. CHAPTER 3. METHODOLOGY 3.1 DATA COLLECTION: INTERVIEWS 3.1.1 Interviewees 3.1.2 Interview questions. 3.2 DATA COLLECTION: OTHER RESOURCES 3.3 CATEGORIZATION SCHEME 3.3.1 Interview data as interactional 3.3.2 Data categorization. CHAPTER 4. LANGUAGE IDEOLOGY IN DIRECT RESPONSES 4.1 APPRAISAL OF ENGLISH. 27 27 29 32 33 34 35 37 38 38. 4.1.1 English as a necessity. 4.2.1 Self-deprecation. 50 59 60. 4.2.2 Deprecation of others. 64. 4.1.2 Positive attitudes toward English. 4.2 DEPRECATIONS OF INCOMPETENCE. vii.

(10) CHAPTER 5. LANGUAGE IDEOLOGY IN SEMI-DIRECT RESPONSES. 73. 5.1 STANDARD LANGUAGE IDEOLOGY 5.2 ACCOMMODATION: ENGLISH IN PUBLIC CONTEXTS. 73 80 81. 5.2.1 Accommodation to English by Non-English Users 5.2.2 Accommodation to English by English-users. 5.3 ACCOMMODATION: ENGLISH IN PRIVATE CONTEXTS 5.3.1 Accommodation by Non-English users 5.3.2 Accommodation by English Users. 84 88 88 92 104. CHAPTER 6. DISCUSSION 6.1 COMPARISONS WITH QUANTITATIVE ANALYSES 6.1.1 Connections between English competence and education 6.1.2 Exchange values and use value 6.1.3 English as stylization. 6.2 WORLD ENGLISHES AND LINGUISTIC OWNERSHIP 6.3 SYMBOLIC POWER OF ENGLISH 6.3.1 English as linguistic capital 6.3.2 English as a marker 6.3.3 Public market values and private market values. 6.4 INTERVIEWS IN DATA COLLECTION CHAPTER 7. CONCLUSION 7.1 SUMMARY AND RESPONSES TO RESEARCH QUESTIONS 7.1.1 Language ideology of English as a necessity 7.1.2 Language ideology in semi-direct responses 7.1.3 Proficiency as a variable 7.1.4 Discrepancies of language ideology in direct and. 105 106 109 112 117 120 121 123 124 125 128 128 129 130 131 132. semi-direct responses. 7.2 DIRECTIONS FOR FUTURE RESEARCH References. 132 133. Appendix 1 – Interviewees. 142. Appendix 2 – Interview Outline (Chinese & English) Appendix 3 – News article (Chinese) Appendix 4 – News article (Chinese. 144 148 149. Appendix 5 – Announcement from Ministry of Examination Appendix 6 – Thread on BBS (Chinese). 150 151. viii.

(11) CHAPTER 1 INTRODUCTION Globalization is mostly comprehended as a cover term of Westernization, or Americanization (Marling 2006), for globalization has led to the redistribution of power and orientation to Western dominance in economy, culture, politics, and most of all, language (e.g. Held et al. 1999; Scheuerman 2008). The increasing contact of individuals from different language backgrounds has brought forth a lingua franca. Western dominance has driven English to become the lingua franca, which acts as a filter and privileges its speakers with the access to global economy. English is hence commonly considered to be an asset of ultimate importance in non-English speaking countries in this globalized era. New forms of economic practices have also altered the allocation of functions of language. Originally serving as a lingua franca in international trade, English has transformed from a mutual code to a global language whose functions are no longer confined to solely meet communication needs. It is well-established that English is deemed to be strongly positive and recognized as a tool of upward mobility in Taiwan. The goal of the present study is to take a step further to investigate how English is conceptualized and what images English projects in Taiwan. To analyze conceptualizations of English, I would like to bring out the focuses of the study with a Facebook message I received from my high school classmate during the study was conducted. In the message presented in (1), she described her encounter with a foreigner at a café.. 1.

(12) (1). Dear Ann, something that happened this morning made me want to strangle myself. A foreigner asked me to translate what he said to a clerk. Why did he turn to me? Because he accidentally cut in line in front of me. I said, smiling, that it was ok and gestured him to order first. He probably misunderstood that I spoke good English. Then he started babbling about how he hated it when clerks placed their hands on cup lids brewing his coffee. I thought he was just trying to banter with me, so I didn’t pay too much attention to what he was saying. And then, he asked me to translate what he said to the clerk. I was caught totally off guard, asking him to repeat. But I was simply too nervous. I didn’t realize that I missed the points entirely until I left. Gosh. I want to strangle myself. I disgraced my school and myself. Do you know that there was a long line after us? Gosh. I can’t believe that I only realized that I completely missed out the most important part.. As can be observed in (1), she revealed the beliefs that it was shameful and condemnable for her to fail the conversation and that she thought she was responsible for the failure. First, the self-initiated interrogative pair in lines 2 and 3 weisheme hui zhaoshang wo ne yinwei ta buxiaoxin cha le wo de dui ‘Why did he turn to me? Because he accidentally cut in line right in front of me’ suggests her clarification that the encounter had nothing to do with her English skills. She was implicating that she was never the right person to talk to when English is involved. Additionally, the phrase zhaoshang ‘turn to’ indicate that the encounter was far from a happy one as zhaoshang entails the passive stance she positioned herself throughout the entire conversation. Besides, zhaoshang usually refers to unpleasant situations which inflict on individuals. Consequently, the interrogative pair in line 2 foretold that she 2.

(13) perceived the encounter to be nasty. Moreover, she claimed that she was wu hui ‘mistaken’ to be a good English speaker (line 4), deprecating herself again as an incompetent English speaker. The severe self-accusation in line 8 diu le xuexiao de lian wo ye diu si lian le ‘I disgraced my school and myself’ implies that she thought it obligated to deal with the conversation with ease, but she failed. Furthermore, the fact that she addressed me by my English name, rarely known and thus let alone used among my high school classmates, could be interpreted as her attempts to either reduce the imaginary distance between her and English or recognize my identity as an English major. Based on her subjective narration of the encounter, it could be scrutinized and concluded that she conceptualized English as a necessity and a responsibility. It could be postulated that she was not the only person to bear these beliefs about English as a necessity and self-deprecation. English has been taught as a mandated subject in formal education in Taiwan for long; moreover, the age at which English is learned as a required subject continuously shifts lower every few year. All these efforts attend to the necessitation of English. Yet, the population of proficient users of English remains small, as Chen (2010) and Tsai (2010) claim. It can be discerned from (1) that English is perceived to be a necessity though not everyone possesses, or believes himself or herself to possess, basic competence. English in Taiwan is deemed to have values that go far beyond fulfilling communication needs. It has been extensively explored and confirmed (e.g. Chen 2010; Lan 2003; Lee 2008; Lin 2003; Tsai 2010; Wang 2000) that English is viewed decidedly positive in Taiwan. Impositions of language planning, media, and Western popular culture project mostly positive images of English and thus massively influence general public’s affirmative stances on English. One important and interesting aspect in (1), in addition to the aforementioned ones, is Ophelia’s relatively 3.

(14) less emphasis on the collapse of information transmission itself. Neglecting how the other party reacted to the situation on the spot, she described the result of the incident with simply one sentence and one reason in line 7 dan yinwei wo tai jinzhang le wo likai shihou faxian wo meiyou shuochu zhongdian ‘But I was simply too nervous. I didn’t realize that I missed the points entirely until I left.’ Instead, a greater emphasis was on her self-deprecation and condemnation of disgracing herself and her school, showing that implications behind failing the conversation were more distressing than the failure of conversation itself. She further remarked that her incompetence in English failed not only the conversation but also her as a person. At the communication level, she believed that she should be responsible for the failure. And at the social level, the message conveys the fact that English is symbolic and indexical to other social attributes. The inability to carry out a successful English conversation means a lot more than the linguistic incompetence in English. The fact that English has obtained and maintained since then its global status is not attributable entirely to its function as a lingua franca. In an increasingly globalized world, the division of labor and the elements of economic activities are redefined owing to expansions of markets from local ones to more integrated and Western-oriented ones. Abstract entities, such as knowledge, service, and most importantly, language, become marketable commodities with exchange values. As termed the knowledge economy by Heller (2003, 2010), new forms of consumer economy centralizes language and English is endowed with economic and symbolic values. The hegemony of English has been credited to its sought-after economic and symbolic values. Unarguably, English is comprehensively considered to be positive, necessary, and instrumental to social upward mobility in Taiwan based on the known literature. The sociolinguistic profile of English is usually thought to be strongly associated with 4.

(15) globalization. However, although many studies have provided detailed accounts on the positive attitudes toward English in Taiwan, to the best of my knowledge, little attention has been paid to the following facets. First, although general attitudes toward English are notably positive, the ideological aspects that shape the attitudes are seldom accounted for. Dyers and Abongdia (2010) suggest that language ideology, mostly known as conceptualizations of language rooted in our social experiences, precedes language attitudes. Accordingly, it is through language ideology that language attitudes can be accounted for. Second, studies on language attitudes have mainly focused on English learners at various levels. Seldom did studies recruit informants of non-English users out of classroom contexts, hindering a more panoramic view on perceptions of English. Third, studies on attitudes toward English mostly collected data through questionnaires and interview, and yet, explicitly self-reported attitudes and perceptions in response to questions may not necessarily present interviewees’ actual conceptualizations of English. Interview data cannot be analyzed as linear. Rather, the interview’s control and manipulation to interviews may affect the reliability of data. Besides, interviewees may reveal different perceptions between answering directly to interview questions and narrating an event related to the interview questions. Conceptualizations elicited in direct response are overt and verbalized. Thus, it could be expected that these conceptualizations may be more close to established, prescriptive social norms while those revealed in semi-direct responses may be more contrastive and context-specific. Consequently, the study identifies the interviewer’s control on data and categorizes data accordingly. The present study is aimed at investigating, qualitatively, language ideology of English in Taipei areas by analyzing both direct and semi-direct responses. The study collects data through interview with 38 informants of various social backgrounds. The thesis focuses on 5.

(16) (1) language ideology of English in direct responses to interview questions, (2) language ideology of English in semi-direct responses (3) proficiency as a variable that influence language ideology of English, and (4) conflicts between general language ideology of English and the one disclosed in interaction. The study is outlined as follows. Chapter 2 outlines a general discussion on language ideology, values of language, and globalization and the rise of English. The sociolinguistic profile of English in Taiwan is also reviewed. Chapter 3 illustrates the methodology and the categorization scheme for the analysis. Chapter 4 discusses language ideology of English found in informants’ direct responses to the interview questions whereas Chapter 5 examines language ideology of English in semi-directed, less interviewer-controlled responses. The findings are compared with quantitative studies and theoretically accounted for in Chapter 6. The concluding remarks, along with the limitations, are presented in Chapter 7.. 6.

(17) CHAPTER 2 LITERATURE REVIEW The study investigates language ideology of English in Taiwan through interviews. Globalization bestows English with economic and symbolic values due to first the need for a mutual code in international commerce and later Western dominance. These economic and symbolic values of English prevail over its practical function as a mutual code, and result in the appropriation of English in non-English speaking countries like Taiwan. The appropriation of English is usually reinforced through institutionalized implementations such as language planning and fostered an internalized social norm. It has been well-researched that English projects strongly positive images and is a highly desirable asset in Taiwan. Owing to both the prescriptive nature of the social norms and influences of methods in data collection, the study takes into consideration the interviewer’s control over interviews. The proximity of English has been discussed with quantitative data on perceptions and attitudes. Nonetheless, as mentioned in the introduction, language ideology, commonly overlooked in studies on language attitudes, provides a theoretical account for language attitudes and perceptions. The review is accordingly divided into four parts. The theoretical framework of language ideology is reviewed in 2.1. Section 2.2 discusses economic and symbolic values of language. The hegemony of English and its relations with globalization are addressed in Section 2.3. Section 2.4 focuses on the sociolinguistic profile of English in Taiwan. Each part itself invites such an extensive body of literature that it is impossible to give a thorough review. Only studies which are of immediate relevance are overviewed. 7.

(18) 2.1 LANGUAGE IDEOLOGY Language ideology can be generally understood as conceptualizations of language which rationalize linguistic practices in social contexts. Therefore, language ideology is unstable and fluid. Globalization has converted the essence of economic activities and made abstract entities like knowledge and language marketable in exchange for profits. What globalization has also altered, in addition to functions of language, are individuals’ conceptualization about what language can do, and what individuals can do to, and with, language. Individuals’ conceptualizations also keep evolving.. 2.1.1 Outline of Language Ideology. Language ideology as a field of inquiry has been approached from various facets so as to bridge up language and society (Woolard 1992). Without a unified definition of language ideology, studies in language ideology differ greatly from one another. The most cited definitions are among Silverstein’s (1979), Irvine’s (1989), Rumsey’s (1990), and Woolard’s (1998). Silverstein (1979) suggests that language ideologies are “sets of beliefs about language articulated by users as a rationalization or justification of perceived language structure and use” (193). Irvine states that ideology is “the cultural (or subcultural) system of ideas about social differentiation that would claim to predict the one from the other” (1989:255). Rumsey defines language ideology as “shared bodies of commonsense notions about the nature of language in the world” (1990:346). For Woolard (1998), language ideology is “representations … that construe the intersection of language and human beings in the social world” (1998:03). Comparing the four definitions, Silverstein puts great emphasis on individuals’ linguistic awareness while Irvine centralizes the functions of 8.

(19) ideology in social differentiation. Rumsey accentuates the sharedness of language ideology and Woolard brings out the essentials of language ideology as representations. These approaches to language ideology differ in dimensions they choose to zoom in. What is mostly agreed upon is that language ideology reflects individuals’ experiences as members of various social groups (Kroskrity 2000; Woolard 1992; Woolard & Schieffelin 1994). Put differently, individuals’ memberships in social groups foster how they conceptualize language. Social divisions and different social experiences lead to multiple language ideologies in a given society. Other prototypical features of ideology are summarized in Woolard (1992). First, language ideology is thought to be conceptual. Silverstein describes language ideology as beliefs, Irvine ideas, and Rumsey commonsense. According to Woolard (1992), the ideational nature of language ideology is deviant from consciousness. Yet the degree to consciousness of ideology is still contestable. For example, Silverstein’s definition of language ideology exhibits the emphasis on individuals’ linguistic awareness while Irvine does not acknowledge the mediating role of consciousness. Second, ideology is power-laden. Language ideology can be manipulated as a tool by dominant groups for its own legitimacy. Language itself is the site where the interests are legitimated and reproduced (Kroskrity 2000). Thompson (1984) proposes that ideology can be exploited to retain the imbalanced power distribution in a society. Moreover, ideology is a rationalization about language, the extreme cases of which involve distortions. Therefore, language ideology as a representation and conceptualization of language contradicts with truth (Cameron 2003). As Woolard (1998) suggests, language ideology is a rationalization that interpret linguistic practices in social contexts. Remarkably, ideology not merely shapes the patterns of talk, but also regularizes (Silverstein 1979, 1985) and neutralizes them (Spitulnik 1998). 9.

(20) 2.1.2 Debated issues of language ideology. In addition to the lack of a unified definition of language ideology, studies in language ideology also debate over alternate stances on values. Studies in language ideology mostly take on two binary perspectives, as discussed in Woolard (1994). Neutral values of language ideology look into cultural representations of language and focus on anthropological aspects. For example, Rumsey (1990) investigates relations between linguistic structures and speakers’ shared notions about these structures. Critical values, on the other hand, account for relations among language, meanings, and power. The critical views of language ideology are mostly adopted as the theoretical framework in studies on language ideology and power. Another debate over language ideology centers on degrees of consciousness of language. Kroskrity (2000) claims that language ideology is normally “tacit” and “very rarely brought up to the level of discursive consciousness” (506). Language ideology, in Kroskrity’s term, corresponds largely to social norms which are embedded in interaction. Though language ideology can be demonstrated in metapragmatic commentary in discourse, it is not the only site where language ideology is located and reproduced. Ideology can be located at various discourse levels. Woolard (1998) proposes that. language. ideology. can. be. situated. in. linguistic. practice. itself,. metalinguistic/metalinguistic discourse, which refers to explicit talk about language, and implicit metapragmatics, defined as “linguistic signaling that is part of the stream of language in process and that simultaneously indicates how to interpret that language-in-use” (1998:09). That is, linguistic items are both parts of language use and criteria of interpreting language use. Kristiansen (2010) identifies two levels of language ideology. Overt language ideologies refer to the conceptualizations of 10.

(21) official values of language in public discourse whereas covert language ideologies are the rationalizations of language values divergent from official values. Wortham (2001) also identifies language ideology at the macro-level and the micro-level with the former referring to a larger and shared belief system in a speech community and the latter the implicit mechanisms adopted by individuals to interpret discourse. Diverse viewpoints on degrees of awareness of language ideology may result in various, and even controversial, accounts for relations between linguistic practices and social phenomena (Woolard 1992). In addition to multi-sitedness of language ideology, individuals may also exhibit various degrees of consciousness of language ideology (Kroskrity 2000). The varying degrees of language ideology in turn influence their linguistic practices, social beliefs and the connections of the two.. 2.1.3. Language ideology and other terms. Language ideology, as shown in the previous section, needs be viewed as a cluster of concepts, rather than a single dimension. However, a lack of a unified definition makes it challenging to pinpoint what language ideology is. Language ideology frequently occurs with terms such perceptions, language attitudes, and beliefs. These related notions blur the borderline among all these terms. The section summarizes comparisons and discussions of these terms. Language attitudes in the literature are mostly analyzed with regards to its three components (Garrett 2010). The cognitive aspect of attitudes refers to fostered associations between our judgments and objects in the social world. The affective component of attitudes refers to emotions and feelings evoked by objects. The behavioral component of attitudes involves how we act in ways that correspond to cognitive and affective components. The validity of the three-component model in 11.

(22) accounting for language attitudes is usually challenged. For one, it is difficult to explore the interconnectedness of these three components. Moreover, the three-component model assumes the congruity of the components and behaviors. Yet, the alignment is denied and challenged. Additionally, it may fall into circularity to determine the relations between attitudes and behaviors. Comparing language ideology and language attitudes, Dyers & Abongdia (2010) make clear that studies in language attitudes mostly emphasize the affect component. Affects shared by both language attitudes and language ideology make the borderline of the two fuzzy. Language ideology is rooted in social experiences of our social positions which involve socioeconomic divisions; consequently, language ideology is shared by groups (Cameron 2003; Woolard & Schieffelin 1994). Language ideology as rationalizations is structured in social-historical events through time whereas attitudes are relatively unconscious and personal (Bucholtz & Hall 2005; Myers-Scotton 2006). As a result, language ideology is a social construct (Dyers & Abongdia 2010; Woolard 1994) while attitudes are mental ones. The most striking contrast of the two is that language ideology shapes language attitudes. That is, language ideology precedes language attitudes (Dyers & Abongdia 2010). Language ideology has been defined by Silverstein (1979) as “sets of beliefs about language.” The conceptual nature of both language ideology and beliefs makes them seem interchangeable. Despite the significance of Silverstein’s definition which leads to the equivalence of language ideology and beliefs, the differences of the two need be identified. Some studies in language ideology are reluctant to define language ideology as beliefs for two reasons (Cameron 2003). Language ideology as rationalization is not objectionable because it is “opposed to truth” (Cameron 2003). On the other hand, beliefs can be either true or false (Talbot 2000). Therefore, belief systems, similar to language attitudes, are mental constructs while ideology is social. 12.

(23) Language ideology in globalization transforms attitudes toward both indigenous languages and English. Language is regarded as commodities with marketable values (Ricento 2010). The new form of economy constructs niche markets in which symbolic values of objects can be commodified (Heller 2003). This new type of economic activity centralizes language as both a resource and a commodity and alters conceptualizations of language.. 2.2 THE ECONOMY OF LANGUAGE. It has been discussed in Introduction that competence in English does not refer to merely the ability to speak the language. In other words, competence in English is bestowed values in addition to communication functions. The section reviews how language is the site at which symbolic and economic values are deposited and power is at work. Language receives values through price formation. Price formation of language developed by Bourdieu (1984, 1986, 1990) assimilates the process of language domination to that of price tagging of merchandise in commerce. Price formation refers to processes in which language is legitimated by linguistic markets. Speakers could access to both material and symbolic profits which accompany language by learning it. Forms which are legitimated are usually spoken by a small number of speakers who, however, control the majority of interests and the operation of markets. As language as interests is unevenly distributed it is able to index to social distinctions. Capital is accumulated and profitable force that transforms for reproduction and enables individuals to obtain social energy (Bourdieu 1986). Thus, patterns of capital distribution depict social constraints and regularities. Bourdieu (1986) states that 13.

(24) capital fundamentally exists in three states. Economic capital, the basic form of capital, is labor that can be directly translated into monetary interests. Social capital is potential resource linked to one’s social networks. Cultural capital is identified in three states, the embodied state (long disposition), the objectification state (cultural goods) and the institutionalized state (sanctions and qualifications). Patterns of capital are a silhouette of social orders. When language as linguistic capital is thought of as economic energy in market, the operation of markets cannot be viewed as autonomous. Irvine (1989) suggests that symbolic and economic values of language need be accounted for in terms of political economy. The linkage between linguistic phenomena and political economy identified by Irvine are established through propositionality, authentication, commodification, incorporation, and indexicality. Propositionality, which refers to denotations of objects in the material world, guarantees mutual intelligibility and linguistic homogeneity in a speech community and facilitate labor divisions and coordination. Authentication refers to the fact that linguistic systems are inevitably accompanied by authoritative statements that legitimate the systems. When individuals buy a language through learning it, they also internalize authorities affiliated to it. Commodification, as Block (2010) offers a more concrete definition, describes value shifts of language from its use-value to exchange-value. Language converts to a commodity whose functions are redefined as measurable skills. Incorporation describes the fact that linguistic performances, as both results and resources of the social system, are both indexed and indexical. Indexicality concerns phenomena in which linguistic diversities are culturally and stereotypically mapped onto social differentiations (Ochs 1992; Silverstein 1985). Indexical values of language are culturally shaped, conventional but not entirely arbitrary (Irvine 1989; Silverstein 2003). The mapping of linguistic variations to social differentiations presupposes a 14.

(25) learned schema of what appropriate language use is. Registers, accents, and practically all linguistic performances are available to activate associations between language and sociocultural divisions. Variables endowed with indexical values can always be reinterpreted and render new social connotations. Consequently, language is never neutral (Bourdieu 1991; Silverstein 2003; Woolard 1998). 2.3 GLOBALIZATION AND THE RISE OF ENGLISH Globalization, which is hardly a new phenomenon, has altered and is continuously altering language in such a dramatic way that language under globalization has received much academic attention. The section provides a general description of globalization and the role English plays.. 2.3.1 Globalization and Language. Globalization covers such a wide range of domains that it is difficult to precisely designate what globalization is. Therefore, globalization is mostly defined and comprehended in terms of conversions it has led to. Held et al. (1999) propose that globalization can be perceived as “a process (or set of processes) which embodies a transformation in the spatial organization of social relations and transactions – assessed in terms of their extensity, intensity, velocity, and impact – generating transcontinental or interregional flows and networks of activity, interaction, and the exercise of power” (1999:16). Held et al. consider globalization to be processes of “transformation” of social constructs. Scheuerman (2008) offers an elaborated perspective. “In popular discourse, globalization often functions as little more than a synonym for one or more of the following phenomena: the pursuit of classical liberal (or “free market”) policies in the world economy (“economic liberalization”), the 15.

(26) growing dominance of western (or even American) forms of political, economic, and cultural life (“Westernization” or “Americanization”), the proliferation of new information technologies (the “Internet Revolution”), as well as the notion that humanity stands at the threshold of realizing one single unified community in which major sources of social conflict have vanished (“global integration”).” Sheuerman’s definition of globalization is comprehended in terms of changes in many dimensions caused by globalization which refers to ongoing processes of transformation. As can be observed from Sheuerman’s definition, globalization is nearly equivalent to the orientation toward Western dominance in economy, politics, and culture. Yet, the seemingly unidirectional transformations under globalization do not elicit homogeneity (Skutnabb-Kangas and Phillipson 2010). Rather, globalization leads to multiplicity and inequality. The increasing contact among individuals of different nationalities spotlights the issues of international and intercultural communication. Languages in globalization are evolving continuously with regards to their forms, functions, and, remarkably, values, owing to the constant contact, (Coupland 2010; Mufwene 2010).. 2.3.2 English as a global Language. To fulfill the increasing demands in intercultural communication, individuals of different nationalities, cultures, and linguistic repertoires must agree upon a mutual code, a lingua franca. Theoretically, any language is capable of becoming a lingua franca, and yet the fact that English is popularly recognized as the lingua franca shows that the global status that English enjoys is a consequence of social dynamics. The militarily powerful British Empire and its colonization left this legacy (see e.g., Crystal 2003; Mufwene 2010). Nonetheless, Skutnabb-Kangas and Phillipson (2010) 16.

(27) stress that English would not have maintained its global status if it were not for the economic and political strength of English-speaking countries. Together they have made English what it is today and led to the global appropriation of English. English has become the foremost world language due to its function as a lingua franca (Mufwene 2010), which is defined by Firth (1996:240) as a ‘contact language between persons who share neither a common native tongue nor a common national culture, and for whom English is the chosen foreign language of communication.’ According to Firth, a lingua franca is a language chosen for wider communication. Nevertheless, the phrase is criticized for its decontextualization and the implication of symmetrical communication (Skutnabb-Kangas and Phillipson 2010). English as a lingua franca in interlingual communication can be symmetrical, when all parties are non-natives of English, and asymmetrical when one party is native speakers of English (Ammon 2010). Though the term is neutral in describing the linguistic phenomena of adopting English for wider communication, it overgeneralizes the social complexities and imbalanced power relations among speakers. The appropriation of English in non-Anglophone communities has made English “indigenized” with local linguistic features to fulfill communicative demands (Mufwene 2009; Ricento 2010). Consequently, the indigenization which is an ongoing process alters forms, functions, and values of English by localizing it. World Englishes refers to different varieties of English spoken as a second or foreign language. The term pinpoints that the majority of English speakers globally do not necessarily speak standard English, if standard English can be objectively defined. With English being spoken world-wide, Kachru (1985) labels English speakers in three concentric circles based on patterns of distribution, acquisition, and use of English. Inner Circle countries refer to those in which English is spoken by monolingual natives. In Outer Circle countries, English is bestowed official status 17.

(28) through institutional imposition. English is usually highly immersed in daily conversation in Outer Circle, and thus English is frequently indigenized with local linguistic features (Mufwene 2009; Ricento 2010). In Expanding Circle countries, English serves little functions. Additionally, the number of English users is relatively smaller compared with that in Outer Circle. Taiwan is categorized as one of the Expanding Circle countries by Kachru. As significant as Kachru’s model is, the typology is severely attacked, for its rigidity in particular. The major criticisms against the model have been summarized in Park and Wee (2009). First, the model assumes clear-cut boundaries that neatly divide countries into three groups. Besides, the model also implicates a direct correspondence between the political history of British colonization and the contemporary sociolinguistic profile of English in a country (Bruthiaux 2003). A related criticism by Jenkins (2003) directs at the fallacy that such categorizations treat countries in each circle as homogeneous. In other words, the model overgeneralizes the heterogeneity within each circle. In addition, English appropriation globally is an ongoing process. Patterns of English distribution are also continuously converting. The accountability of the three-circle model will be challenged (Park & Wee 2009). Moreover, the model still labels the distribution of English at the level of nations and ignores social factors such as gender and ethnicity within them (Pennycook 2003). Furthermore, the model focuses on the spread of standard English in more formal domains such as education and business and ignores domains of popular cultures and youth sub-cultures. According to Pennycook (2003, 2010), these domains demonstrate the appropriation of dialectal, variational, and even vernacular English. The most fundamental debate ignited by the model concerns the fact that the model reinforces the native-nonnative dichotomy and reproduces the hegemony of Inner Circle countries and the norms spoken by them. Higgins (2003) adopts Norton’s (1997) 18.

(29) theory of ownership as an alternative to the dichotomy between native speakers and non-native speakers. Though confronted with these criticisms, Kachru’s model is still very prominent in studies on world English.. 2.3.3 English Learning, Social Implications, and Concerns. It has been mentioned that English has sustained its global status due to the economic strength of its speakers. English as a global language, similar to other related terms such as world Englishes, English as a world language, English as a lingua franca, describes a neutral linguistic phenomenon in which speakers adopt English in intercultural communications. However, these terms all reflect social phenomena in which power and the imbalanced distribution are involved. English and its global status alone could not account for the continuous spread of English in non-English speaking countries. Language planning refers to institutionalized and conscious efforts to enforce language learning (e.g. Cooper 1989; Fishman 1974; Haugen 1983). It includes status planning (the positioning of the language in a society), corpus planning (functions of the language), and acquisition planning (the promotion of language through education). Through language planning, a language is officialized, standardized, and maintained. Remarkably, language planning also involves multitudinous social dimensions (Cooper 1989; Tollefson 1991). English has been one of the mandated subjects in formal education in non-English speaking countries. With its global status, English functions as a medium through which individuals participate in global events. For modernization and global competitiveness, English is implemented and integrated into school curricula in non-English speaking countries. In Asian-Pacific countries, ages at which English is 19.

(30) taught as a required subject in compulsory education keep shifting lower every few years (Nunan 2003). The planning and reforms reveal that English is highly valued and emphasized. English is significantly unequally distributed in a society. Due to its instrumental values, English is thought to promise economic and social advancements (e.g. Grin 2001; Park 2011). Competence in English is a marker of social distinctions (Ricento 2010). Nonetheless, the direct association between competence in English and socioeconomic status raises two questions—assessments of English competence and the indeterminancy of the relations between the two. Linguistic competence which refers to the knowledge of proper language use (Hymes 1967) is in fact subjectively defined. Blommaert et al. (2005) assert that competence depends largely on linguistic markets for its recognition. In this fashion, legitimacy of competence is determined by how this language is positioned in a particular regime. Regimes are capable of validating and incapacitating speakers’ repertoires. In other words, competence and the recognition of it are context-specific. Since competence is not objectively measurable, values of the language and what speakers can obtain via competence in English are also determined by linguistic markets (Park 2011). Furthermore, even with the support of empirical evidence that competence in English and its speakers’ socioeconomic status are related, there is a certain degree of difficulty to justify that English indeed has economic value. To begin with, even when competence in English and economy are related, it does not necessarily indicate that they are in causal relations. Additionally, there is a risk of circularity in deciding whether it is language that affects economy or the other way around (Grin 2001; Park 2011). Moreover, a bundle of other determinants are also in positive correlation with earnings. It suggests the inappropriateness to claim that English leads to financial advancements (Grin 2001). The stereotypes, however, are still proven so in Taiwanese 20.

(31) contexts. One concern aroused by the global appraisal of English lies in the threat that it may pose to multilingualism. The number of English speakers is continuously growing. Hence, English is thought to eclipse multilingualism. The view was debated and soon refuted owing to the allocation of functions that English serves and the limited numerical strength in EFL and ESL settings (Phillipson 2003). English was first approached for intercultural communication (Brutt-Griffler 2002) and is acquired for instrumental values. The exploitation of English is goal-oriented and purposeful. For individuals who acquire English as a second or foreign language, English is seldom adopted for identity marking (House 2003). Compared with one’s indigenous languages, English is allocated with different functions. Consequently, the discrepancies of functions of the global language and indigenous language need to be identified. The different allocation of functions rules out the possibility of English waning indigenous language and posing a threat to multilingualism. The fact that English neither replaces nor invalidates local languages is attributable to still relatively limited numerical strength as well. Mufwene (1994, 2008) compares the global status of English to that of Latin in order to debunk the myth that English is a killer language. Latin, similar to English, was only favored as a communication tool in international trade. By the same token, though English enjoys high social prestige in Outer and Expanding circle countries like Taiwan, English is still spoken by a relatively small number of users. English is thus confined to certain domains while indigenous languages, Taiwanese and Chinese, can still fulfill nearly all communication demands (Tsai 2010). Moreover, Crystal (2003) asserts that language domination and death in fact operate independently from the spread of a global language. To sum up, the concern that English threatens multilingualism is invalid for it will not replace one’s native languages. 21.

(32) 2.3.4 Linguistic ownership. The prevalence of English globally has cast spotlights on the concern of linguistic ownership, with nonnative speakers astonishingly outnumbering its native speakers (Crystal 2003). The labeling of native and nonnative dichotomy and Kachru’s (1985) model reinforces the hegemony of Inner Circle norms. A more deliberate account for the legitimacy of English is brought up. Ownership is proposed as an alternative of previous models (Higgins 2003). Widdowson (1994) suggests that authenticity, the ability to demonstrate standard English competence, plays a make-or-break role in ownership recognition. However, Widdowson also describes the deficiency of the strong conventional association between authenticity and ownership in catering to the world trend. The conventional approach, the purist school, emphasizes the uni-standard form of English, as Wee (2002) summarizes. The pragmatist school, on the other hand, advocates pluricentric standards of English and proposes that English is owned by all its users. The study adopts the pragmatists definition of ownership and investigate how ownership of English influences language ideology of it.. 2.4 ENGLISH IN TAIWAN. The current sociolinguistic profile of English in Taiwan results chiefly from various institutional power—education, media, and popular culture. Generally, the number of proficient English speakers is still relatively small even though it is steadily increasing. As English is of high prestige but with a small number of proficient English users, functions of English are mostly perceived to be instrumental (see e.g. Chang 2001; Chen, Warden & Chang 2005; Wang 2000). In fact, the instrumental motivations in fact surpass its practical functions (Chen 2010). 22.

(33) 2.4.1 English education in Taiwan. English education in Taiwan has undergone tremendous changes in the past decade. Though English has always been a compulsory subject in high school for long, it was conceived as insufficient to equip every student with proficiency for obvious reasons. Along with other prospects—developments of international perspectives, maximizations of effects of the critical period, catering to parents’ expectations, and longer immersions in English (MOE 1998), the demands pushed forward the reform of the new language policy. Chen (2010) stresses that the new policy enacted in 2001 differed from the previous one in its focuses on earlier exposure to English at the elementary level, English as the language of instruction in higher education, and basic competence for all citizens (Chen 2010). The reform of the policy put great emphasis on making English as a medium of information rather than as merely a subject. It has been affirmed in the previous sections that language planning involves social dimensions (Cooper 1989; Tollefson 1991) owing to the fact that language planning endorsed by the institutional authority can affect speakers’ language ideology of English. English is therefore mostly viewed as positive in public sectors. The new focus on earlier exposure to English is particularly interesting and controversial. The situation is worth the attention for two reasons. For one, the parent-directed policy is inclined to load policy implementers and the government with much pressure (Lin 2006; Oladejo 2006). For another, although this reform reflects and meets expectations of general public, parents’ expectations on early exposure to English contradict with the government’s motives for the reform. English in global contexts is a medium of international communication and an access to foreign information. However, Lee (2008) asserts that parents are eager to have their pre-school children learn English for, again, its instrumental functions. Though policy 23.

(34) implementers also consider plausible teaching English at the elementary level (Duo 2003) and the early exposure to English has received positive feedbacks from all parties, they apparently do not see eye to eye with each other on reasons of English necessitation. Instrumental values of English have been an effective motivation for English learning in Taiwan (Chen 2006). The new policy is also aimed at developing English to be a medium of information access by advocating English as the language of instruction in higher education. Beliefs of and attitudes toward English by highly educated students may be diverse. Notions of integrative and instrumental motivations of English start to kick in as English is no longer a mandated subject. It is found by Chou (2005) that proficiency levels are prominent in shaping attitudes towards English. In addition, for proficient speakers like senior English majors, English can even serve integrative functions as an ingroup marker due to their long-term immersion in English. The finding suggests that proficiency levels may indeed plays an essential role in shaping conceptualizations of English. The focus on attitudes toward English held by general citizens is the least explored in the literature. Seldom did studies elaborate on general public’s attitudes toward and perspectives of English as most of the studies have focused on either English learners, such as students, or English users with various levels of proficiency. To head for the prospect of being an international country, service industries at the national counters such as airports is among the first group affected by the reform. In 2005, the government planned to stipulate that taxi drivers at airports to take the basic level of General English Proficiency Test (GEPT henceforth). The news took wide media coverage on January 7, 2005. The next day on January 8, 2005, the editorials were crammed with both voices of agreements or discontendedness toward the reform. In one report, Li (2005) accounts for taxi-drivers’ collective of the reform and 24.

(35) explains that English proficiency guarantees higher income and stable customer-hood. It again points to the fact that English is instrumental. To sum up, the three focus of the new policy targeting at different groups of citizens are in fact aiming at making English prevalent in Taiwan. Through language planning, social values and status of English are also imposed on citizens. However, language normality in a given society is never a result of language planning solely. As Blommaert (2009) states, institutions, though playing a primary role in language normality, are never hegemonic.. 2.4.2 Other institutional power. Language ideology can also be produced and reproduced in media (Spitulnik 1998). Take magazines in Taiwan for example, English or English-Chinese code-mixing is a marked language choice which has been manipulated to attract attention. At the same time, English or code-switching vivifies expressions (Shih & Sung 1995). To ensure intelligibility, that is, to make sure that readers can read the English expressions with even limited proficiency, usually short and basic English words, nouns in particular, are preferred (Chen Cheryl Wei-Yu 2006). In press, English projects an image of cosmopolitanism and modernism. Press representations, in particular prestige press, usually allow readers to view the hegemony of English as uncritical and positive (Demont-Heinrich 2008). In the previous two sections, it has been demonstrated that English in Taiwan is mostly treated as instrumental. And owing to power imposed through language planning and economic values of English, English is bestowed symbolic power in the society. Hence, it is not surprising to find that English is exploited by speakers to construct authority. Lan (2003) discusses that the employer-employee hierarchy between Taiwanese employers and Filipina domestic workers can thus be easily 25.

(36) challenged through manipulation of English. Taiwanese employers use economic capital to establish authority while Filipina domestic workers use their cultural capital to display resistance. Ownership of English as cultural capital sometimes outruns the ownership of economic capital. Through interview, Lan claims that Taiwanese employers are self-reported to be less unwilling to hire well-educated domestic workers. Though economic capital is still the determinant in the employment relation, English sometimes has higher prestige in markets in transnational interaction. And the symbolic power of English even precedes its actual use in Taiwan. The sociolinguistic profile of English in Taiwan is strongly related to language planning and media. Through implementation, English in Taiwan is viewed strongly positive and facilitative to upward mobility.. 26.

(37) CHAPTER 3 METHODOLOGY To investigate language ideology of English in Taiwan, the study adopts a qualitative approach and has collected data chiefly through interviews. The sociolinguistic profile of English in Taiwan was mostly conducted in quantitative analyses (e.g. Wang 2000; Chen 2010). Moreover, the issue was usually approached from the facets of English education in Taiwan (e.g. Chang 2008; Hsieh 2006; Lee 2008; Oladejo 2006). Relatively little attention has been paid to the social dimension excerpt for language attitudes toward English. The study is aimed at developing a sociolinguistic account for conceptualizations of English. Gathering data through interviews allow observations of rationalizations of English. Yet, the drawbacks of conducting interviews, such as the interviewer’s control in interviews and interviewees’ awareness and pressure of providing information conforming to shared social norms, are taken into consideration. Consequently, the interview data contains such rich information that they cannot be treated as flat and regular question-and-answer sequences.. 3.1 DATA COLLECTION: INTERVIEWS. The section describes in detail the recruitment of interviewees and the outline of the interview questions.. 3.1.1 Interviewees. In order to investigate how English is conceptualized, the study interviewed 38 persons in Taipei area (Taipei City and New Taipei City). Studies in language 27.

(38) ideology of English in Taiwan are relatively scarce and studies on language attitudes toward English seldom recruited informants who with relatively limited command in English. To look into how linguistic ownership of English influences language ideology, the study has interviewed both English users and non-English users. Since competence is gradable, it is hardly plausible to define competence and incompetence without a gray area in between. In other words, how competent one can be categorized as a competent English user and how incompetent one is regarded as an incompetent English user are ambiguous. To avoid the gray area, the study recruited informants at the two ends of competence continuum and magnified the informants’ contrasts in terms of their competence in English. English speaking users are defined in this study as those who make a living with their English competence, use English on a daily basis, or score higher than 80 in iBT. On the other hand, non-English users refer to those who do not display any of these three features. Among the 38 informants, twenty four non-English users and fourteen proficient English users participated in interviews. To gather as diverse information as possible, the study carefully recruited informants of various social backgrounds. The 38 interviewees aged from 19 to 51, with their occupations and education levels varying. Thirty four interviewees are native speakers of either Taiwanese or Chinese. Four are Hakka, but only three of them speak Hakka with native fluency. To avoid the biased analyses which recruiting informants from my own social networks may result in, informants were asked to bring interviewees unknown to me. By doing so, more different information was obtained due to informants’ different backgrounds. Besides, the participation of both the acquainted and the unacquainted could prevent the awkwardness and unnaturalness that interviewing total strangers may cause. The presence of familiar individuals to both the interviewer and interviewees generate a more relaxed atmosphere and urge other interviewees to open up more. The list of 28.

(39) interviewees participating in the study is in Appendix 1. Face-to-face interviews were audio-recorded and notes were taken during interviews. All interviewees were conscious of the presence of the recorder and gave consent to the study to use the information during their interviews. Before each interview began, a free banter was set to diminish interviewees’ anxiety caused by the presence of the recorder and the coming interview. Afterwards, the audio-taped interviews were overviewed again with more notes taken before they were partly transcribed based on the notes. All the interviewees volunteered in interviews and did not receive any monetary rewards for participation. All the informants were current residence of either Taipei city or New Taipei City when interviews were conducted. Language ideology held by residence in Taipei undoubtedly differs from the counterpart held by residence in other cities. The geographical confinement is surely a limitation of the study.. 3.1.2 Interview questions. The total length of the interviews is 24 hours and 18 minutes. The interview contains three sections of questions, personal information, daily use of English, and attitudinal perceptions and conceptualization of English. An outline of the interview questions is in Appendix 2 in both Chinese and English. The three sections of interview questions were repetitively revised during the data collection, so as the ways in how these questions were phrased. Interviewees completed their personal information in the first section, including ages, occupations, native languages, and language repertoires. The interviewees were also inquired about their self-ascribed English proficiency levels, and standardized English assessments they had taken. Inquiring both self-ascribed competence in 29.

(40) English and information about standardized assessments is to avoid the following two situations. First, self-ascribed proficiency allowed observations of interviewees’ stances on learning English, as they habitually slipped in evaluative comments on their proficiency such as self-deprecation. The information allows a close-up examination of how English is viewed and how informants evaluate and position themselves with regard to their English competence. Second, it had been expected that some informants, non-English users in particular, may not have taken any proficiency assessments. The information of their proficiency levels would have been possibly left unanswered if interviewees had not self-reported their competence in English. In addition to proficiency levels, informants were also asked to recall their experiences in English learning so as to look into reasons and values of English appropriation in Taiwan. In the second part, the questions were designed to observe interviewees’ daily contact with English. Interviewees were first inquired to comment on their competence/incompetence in English. Interviewees were asked to recall their experiences of using English, talking to foreigners, and job hunting. The purposes were to both obtain information about English use and scrutinize underlying rationalizations and perceptions of English via the informants’ responses. Take the informants’ narrations of talking to foreigners as an example, the initiative in conversation, code choices, and the initiative in speaking English all index to how English is conceptualized. Job hunting experiences, on the other hand, revealed interviewees’ beliefs of roles which English plays in labor market in Taiwan. This is also the section in which discrepancies between English users and non-English users started to display. Non-English users withdrew and held themselves back to such a great extent that the length of non-English users’ interview in the section was significantly shorter than that of English users’. 30.

(41) In the last section, the interview questions were directed to acquire information regarding the interviewees’ perceptions and conceptualizations of English. Open-ended questions concern controversial issues that could evoke deeper discussions. The interviewees were presented one current news article, one announcement from Ministry of Examination and one post from Bulletin Board System (henceforth BBS). Each of the issues presented to the informants contains a set of questions, and the interviewees were asked to elaborate their views on these issues. The goals, however, were not merely to seek whether the informants voted for or against these current issues, but to look into, again, conceptualizations of English. The news article on November 2, 2011 [Appendix 3] regards the government’s prohibition against cram schools and bilingual kindergartens for pre-school toddlers. The government was attempting to direct the false emphasis on early start and overlearning in English acquisition. The news hit the headline again in early June, 2012, as the regulation would be enacted in summer 2012 [Appendix 4]. The interviewees were asked whether they argued for or against the new regulation. An extended question is regarding their plans for their kids in terms English education. During the data collection which lasted for a relatively long span, some informants were presented with the earlier article but some were shown the updated one. The announcement from Ministry of Examination regards the government’s concern to integrate English into national examinations [Appendix 5]. Due to the length of the official announcement, the main idea was summarized orally during interviews instead. The BBS thread, taken from the Gossip forum, is a debate originated from a codemixed English word in a Chinese post [Appendix 6]. The codemixed ‘offer’ in the Chinese post was accused of showing off by a person. Irritating a large crowd, the person was soon found out to be a college student through human flesh search. The 31.

(42) college student was severely attacked owing to the fact that others took his harsh comment as an indicator of his incompetence in English. The thread was quickly filled with taunts at that college student. It was possible that the college student did read English and that people knew that the student did not cause a stir simply due to his incompetence. That is, both parties were aware of the fact that incompetence may not be the only reason why codeswitching is disprefered. The thread was expected to evoke deeper discussion about the issues investigated in the study as it contains stereotypes of both social implications of English and the correlation between English and education levels. The interviewees were presented the edited version due to its originally 19-paged length which was indigestible during interviews. The interviewees were asked whether they argued for or against these current issues. The purposes were not merely to seek their opinions on the issues, but to analyze their language ideology of English through their descriptions.. 3.2 DATA COLLECTION: OTHER RESOURCES. Interviews were the major resource of data to investigate language ideology of English. However, the data may be insufficient in yielding a well-developed analysis for the following reasons. First, although the study recruited interviewees with varying backgrounds and different levels of English competence, still it remains a relatively small-scaled study with limited informants and geographical restriction. Moreover, the numbers of informants in each social group were imbalanced. Among the 38 informants, thirty one have bachelor degrees or even higher while seven interviewees completed their high school education. There were only 7 males but 31 females. Three informants were under twenty, eighteen informants in their twenties, 13 informants in their thirties and 4 informants in their forties or above. The 32.

(43) informants were unevenly distributed in terms of social factors, including competence, education, gender, and age. As a corollary, it is prone to yield a biased analysis. Studies on language attitudes toward English as a field of inquiry were usually targeted at either English learners at various levels or students in schools. Therefore, conceptualizations of language by citizens from all walks of life, for example, those in service industry, are less explored. Relying on interview data for generalizations is inclined to result in biased analyses due to the lack of references. These concerns show that the interview data are insufficient to develop a sound analysis in this study. Consequently, the study also selectively collected interesting editorials, news articles and messages from online social networks such as Facebook and BBS. These materials were approached with the theoretical frameworks of critical discourse analysis.. 3.3 CATEGORIZATION SCHEME. The section outlines a description of how data are analyzed in the study. Yet, the analysis needs to start with identifying what interview data are. Interview data, similar to naturally occurring data, are also conjectures between appropriate language competence and social positioning. Interview data are therefore interactional even though they are usually treated as artificial. Interview data are interactional in that they reveal the social positioning of both interviewers and interviewees. Contents and flows of interviews are manipulated by interviews, so interviewers’ control affect degrees of how interactive interviews are. Consequently, the influences of interactions and interviews as one type of social events are acknowledged in the study.. 33.