System 53 (2015) 60e72

Contents lists available at ScienceDirect

System

journal homepage: www.elsevier .com/locate/system

Teaching English as an international language in China:Investigating university teachers' and students' attitudestowards China English

Weihong Wang a, b, *

a China University of Geosciences, Wuhan, Chinab The University of Hong Kong, Hong Kong

a r t i c l e i n f o

Article history:Received 8 December 2014Received in revised form 7 June 2015Accepted 12 June 2015Available online 26 July 2015

Keywords:China EnglishEnglish learnersEnglish teachersLanguage attitudes

* RM#109, Hui Oi Chow Science Building, PokfulaE-mail address: wangwhw@connect.hku.hk.

http://dx.doi.org/10.1016/j.system.2015.06.0080346-251X/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

This study investigated the attitudes of Chinese university students and teachers towardsChina English (CE), an emerging variety in China and the ideological underpinningsbeneath their attitudinal responses. In the study, 1589 university students and 193 Englishteachers were asked to evaluate the understandability and acceptability of some potentialfeatures of CE with a questionnaire survey. Among them, 31 students and 33 teachers werealso invited to provide reasons for their survey answers. The study revealed that both thestudent and teacher participants were reluctant to accept CE as a pedagogical model buttheir attitudes diverge as specific CE features were involved. The in-depth exploration oftheir justifications identified that the widespread native speaker English ideology andChinglish stigma were more important reasons leading to their negative evaluations of CEthan concerns for the communicativeness of CE to the outside world. Findings of the studymay have important implications for English education in contexts where local varieties ofEnglish are emerging.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

The global spread of English has been reshaping the sociolinguistic realities of the language. It has led to the emergence ofa number of local English varieties in different places of the world (Bolton, 2012; Kachru, 1985). The rise of English for in-ternational communication among people from diverse linguistic and cultural backgrounds has also challenged the domi-nance of native speaker English in English language education (Jenkins, 2006; Ke& Cahyani, 2014). Proposals have beenmadeto include more localised English varieties in the teaching and learning English enterprises alongside the conventionallyadopted standards of British and American English (Kirkpatrick, 2006; Sharifian, 2009). Nevertheless, attitude research haspersistently noted language learners' preference for native speaker English instead of other choices (Timmis, 2002; Young &Walsh, 2010). Language learners were even found to be reluctant to accept the variety emerged in their own contexts (Chan,2013). Therefore, it is necessary to find out what specific about these varieties impairs peoples' willingness to accept them asalternative pedagogical models. To this end, China English (CE) was taken as a case in this mixed-method study to exploreuniversity English learners' and teachers' perceptions of it in the context of China. Given that the number of English learners

m Road, Hong Kong. Tel.: þ852 59839753.

W. Wang / System 53 (2015) 60e72 61

and/or users has well exceeded 400 million in mainland China (hereafter China, Wei & Su, 2012), the investigation may haveimportant implications for the localisation of English education in China and similar contexts.

2. Literature review

2.1. Problematising native speakerism in English language education

Native speaker English, typically British and/or American English, and their speakers have often been endorsed as targetmodels in English teaching and learning practices in many contexts (Cook, 2007; Yoo, 2014). The indiscriminate privilege ofnative speakers and their English contributes to the creation of a pervasive culture of native speakerism in such contexts(Kumaravadivelu, 2003; Waters, 2007). As Holliday (2005) argued, the over-representation of the native-speaker point ofview at the expense of non-native-speaker ones (native speakerism) had “a massive influence and exists to a greater or lesserdegree in the thinking of all ESOL [English to Speakers of Other Languages] educators” (p. 7). It produced “realities ofexclusion, discrimination and rationalizations for intervention and ‘cultural correction’” (Kabel, 2009, p. 17). With thedevelopment of English in global contexts, the dominance of native English in education has been critically contested inrecent years. It has been argued that the adoption of native English as pedagogical models privileges native speakers assanctioned English owners and norm providers, whereas non-native speakers as underachieving learners and emulators(Kirkpatrick, 2006). Power imbalance of such may create considerable inequities between those who haves and those whohave-nots (Nunan, 2003). What's more, to teach English as a native language also goes against the reality that many indig-enous English varieties have already developed to express meanings in locally relevant ways and as identity markers for theirown speakers (Alsagoff, 2010). A native English model fails to attend the diversified realities of English in the global context.Besides, for most learners who are living outside of Anglo-American countries, it is also impossible for them to achieve native-like proficiency by learning English mainly through macroacquisition (Brutt-Griffler, 2002). A native speaker model actuallysets up an unattainable goal for them (McKay, 2010). More critically, a native speaker model does not necessarily guaranteesuccessful communication in the international arenawith the increased participation of speakers from different linguistic andcultural backgrounds (Jenkins, Cogo, & Dewey, 2011).

To address the dominance of native speaker English in language education, a variety of pedagogical proposals have beenmade. Some suggest enhancing learners' linguistic flexibility by exposing them to various English varieties (Morrison &White, 2005; Suzuki, 2011). Others recommend the teaching of international communication strategies to English learners(Seidlhofer, 2011; Sifakis, 2009). Among them, one frequently discussed proposal is to integrate the variety generated inlearners' own local context into education. It is argued that a localised model is culturally, politically and linguistically morerelevant to teachers and learners in that context than a native speaker model (Kirkpatrick, 2006, 2007). It has the potential toincrease learners' ownership of English for it is closer to their own life and can enable them to communicate the culture andvalues of their own country to international audience (McKay, 2010). In spite of such enthusiasm in promoting local Englishvarieties, there are still concerns about the appropriateness of having local varieties as pedagogical models in contexts whereEnglish has restricted presence and has not yet developed into a legitimate variety (Bruthiaux, 2010). However, in countriessuch as Japan, scholars have started calling for “the teaching of [Japanese] English as a de-Anglo-Americanised internationallanguage” (Hino, 2009, p. 107). They argue that although standardised local varieties are not yet formed in many contexts,“nativization indeed takes place […] in away that both reflects and allows users to express their indigenous values” (Matsuda& Friedrich, 2012, p. 20). Attitudinal studies, though have continually documented frontline teachers' and learners' preferenceto native English, have also begun to notice participants' emerging appreciation and attachment to English developing in theirown context (McKenzie, 2008; Sasayama, 2013). More studies are needed to probe into the subtleties and nuances involved inthese attitudinal responses to local English varieties in different contexts.

2.2. China English and English language education in China

In China, English proficiency has long been promoted as indispensable for the nation's modernisation and internation-alisation though English is seldom used for daily communication (Gao, 2012). English language education has been acompulsory course in mainstream education from secondary schools to universities ever since 1978 and further expanded toprimary schools in the new millennium (Lam, 2005). At present, English has enjoyed the largest population of learners incomparison with other foreign languages taught in China (Xu, 2010). Nevertheless, it is mainly native speaker English,particularly American or British English, that has been advocated under the name of ‘standard English’ in the educationalsystem (Gil & Adamson, 2011).

In recent years, China English has been promoted as a developing English variety emerging in the context of China and apossible pedagogical choice for Chinese English learners to challenge the dominance of native speaker English in China(Deterding, 2006; Gao, Liao, & Li, 2014; Hu, 2004; Kirkpatrick, 2007; Xu, 2010). The term China English (CE) was initially putforward by Ge (1980) in translation to refer to English expressions unique to the Chinese culture. However, it had not arousedmuch attention at that time because with the dominance of native speaker English in education, English usages with Chinesecharacteristics but deviating from native speaker norms had been pervasively taken as Chinese English or Chinglishdaninterlanguage that was produced by language learners in their learning process and was somewhere in-between their mothertongue and the target language (Hu, 2004; Wei& Fei, 2003). For reasons of such, the two terms, Chinese English and Chinglish,

W. Wang / System 53 (2015) 60e7262

have strong pejorative implications in China (He & Li, 2009; Henry, 2010). But for advocators of CE, English expressions withChinese characteristics are natural results when English encounters Chinese (Fang, 2008). As Kirkpatrick (2007) argued, withthe fast development of English in China, CE might be an inevitable outcome in the future and the very variety that couldgenuinely empower Chinese learners and users of English:

I argue that although China was an expanding circle country following Kachru's classification, a Chinese variety ofEnglish was developing there and developing faster than has been the case in outer circle countries. In such contexts, itseems inevitable that the local endonormative model will become the one used in classrooms (Kirkpatrick, 2007, p.192).

Xu (2010) joined Kirkpatrick (2007) and elaborated four possible benefits of teaching CE in China: 1) it better addresses thefuture needs of Chinese English learners; 2) it contextualises the learning and teaching experiences of Chinese speakers; 3) itmaximises mother tongue experiences during the learning process; and 4) it raises learners' awareness of their Chineseidentity. Wen (2012) further proposed to reshape English education in China towards the teaching English as an internationallanguage framework and put forward a three-dimensional pedagogic model to improve Chinese students' linguistic, culturaland pragmatic competences within the framework. The learning of local English features and the ability to express localvalues and cultures in English were included as indispensable components in her multi-competence developing model:

For linguistic component, learners are expected to be exposed to native varieties, non-native varieties, and thoselocalized features, which are needed to express the learner's own culture…they are encouraged to learn how todescribe and explain their own cultures in English to speakers from other countries (Wen, 2012, p. 86).

However, one has to admit that by emphasising the Chineseness of CE, it, in essence, acknowledges the influence ofChinese language and culture in the development of CE, just like Chinese English/Chinglish. As a result, the three concepts maynot be neatly separated as explained in Hu (2004):

[T]here is no clear boundary between Chinglish and Chinese English on the one side and China English on the other: itis not possible to place them neatly into two categories. Instead, they are situated on a continuum and progressivelymerge (Hu, 2004, p. 27).

Because of the inseparability of CE, Chinese English and Chinglish, scholars tend to take CE as an emerging variety, or directlycall it Chinese English to highlight the developing nature of English in China (Kirkpatrick, 2007; Xu, 2006, 2010). In this study,China English (CE) is adopted instead of Chinese English, considering the ‘stigma’ attached to the latter. However, these effortsto conceptualise CE point to the potentiality to study the development of CE by investigating people's willingness to alignwithChinglish/Chinese English or CE in the continuum, and their desire to express Chinese cultures and identities while speakingEnglish. As primary English learners and/or users in China, classroom students' and teachers' attitudes deserve attention.

The few attitudinal investigations currently available seem to have revealed a progress in teachers' and students' evalu-ation of CE. Both of the two earlier studies of Kirkpatrick and Xu (2002) and Hu (2004) found university students' over-whelming support of native English, in particular American English, as pedagogical models in language education, and pooridentification with CE. Teachers in Hu (2005) seemed to be relevantly positive. She reported that although nearly 80% of theteachers investigated preferred American English, CE was also accepted by 53% of them. The later study of He and Li (2009) onboth teachers and students identified their ascended support (62.6%) for the incorporation of “select features” of CE in theclassroom (He& Li, 2009, p. 79). But they did not specify what these ‘selected features’would be. Built upon He and Li (2009),He and Zhang (2010) explored possible ‘select features’ by examining CE pronunciation and grammar. They found that overhalf of the participants (55.4%) had no problem with the Chinese accent in speaking English, but 71% of them insisted onstrictly following native speaker English norms of grammar. To conclude, it seems that Chinese university teachers andlearners do not reject CE as a whole.

Since a series of linguistic features of CE have been identified at phonological, lexical, syntactic, and discourse pragmaticlevels in literature (e.g. Bolton, 2003; Deterding, 2006; Kirkpatrick, 2007;Wei& Fei, 2003; Xu, 2006, 2010; Yang, 2005; Zhang,2002), it is possible to expand the investigation of ‘select features’ from CE pronunciation and grammar in He and Zhang(2010) to lexis and discourse pragmatics as well. What's more, Hu's (2004, 2005) separate investigations on students andteachers seem to reveal that the two groups have different attitudes towards CE. There is a very recent study probing teachers'attitudes towards concrete CE features (Yang & Zhang (2015) and revealing their pervasive dislignment with CE, but nosimilar study on students has been available yet. Further research differentiating the two populations is necessary to pindown the attitudinal progress identified in He and Li (2009) and He and Zhang (2010). In addition, previous studies aremainlyitem-response surveys, which could collect data on participants' surface-level responses, but fail to explore the in-depthreasons underpinning their attitudinal responses. To integrate CE in education, a deeper understanding on issues such aswhy participants hold this or that attitude towards CE and how such attitudes are formed might be more enlightening.Therefore, for pedagogical considerations, I framed the study in the following three questions:

(1) How do university teachers and students in China evaluate CE? What are their attitudinal responses towards CEfeatures in the area of phonology, lexis, syntax, and discourse pragmatics respectively?

(2) Are there any differences between teachers and students in their evaluation of CE?

W. Wang / System 53 (2015) 60e72 63

(3) What are the reasons underlying teachers and students' attitudinal responses toward CE?

3. The inquiry

To address these questions, both quantitative and qualitative methods were adopted in the design. First, a survey in-strument based on CE features identified in the literature (see the Appendix) was developed to gather quantitative data on theacceptability of CE to the twomajor populations in educational settings, that is, teachers and students. The questionnaire used18 language examples to elicit university English teachers and learners' attitudes towards CE and the examples covered fourmajor aspects of CE features discussed in the literature (i.e. features in phonology, lexes, syntax and discourse pragmatics). Allof the 18 examples were selected from previous studies as listed in the literature review section, but were revised slightly tomake sure that each example illustrated only one category of CE features. Specific to the 18 examples selected, Item 1 wasdesigned to elicit participants' attitudes towards phonological features of CE, Items 2e7 were lexical features, Items 8e13syntactic features and Items 14e18 discourse pragmatic features. A list of language examples used in the study is alsoenclosed in the Appendix. Phonological features of CE were presented together in one example for that was closer to realcommunication situations where a series of phonological features usually occurred as a whole in a speech stream instead ofseparately presented in different streams.

The questionnaire was administered to 1589 university students (785 males, 802 females, and 2 unspecified) and 193university English teachers (38 males, 154 females, and 1 unspecified) through QQ d an instant messaging software widelyused among people inmainland China. The software provided participants easy access to the audio file needed to evaluate thephonological features of CE in Item 1. The 1782 participants were from 12 universities in Hubei, Shan'xi, Sichuan, Hebei,Shanxi and Henan provinces, two universities from each. The 1589 student participants were from a wide range disciplinesincluding Arts and Designing, Journalism and Communication, Business Management, Economics, Engineering, ComputerSciences, Geology, Medicine, and Biology. Their length of formal English learning ranges from 6 to 15 years, with an average of10.7 years. For the 193 teacher participants, their length of English teaching ranges from 2 to 37 years with an average of 8.6years. The participants were invited to evaluate CE bymaking choices for the 18 language items in the questionnaire. For eachitem, two dimensions of attitudes were to make: understandability (1 ¼ Yes, I understand it, 2 ¼ No, I don't understand it) andacceptability (1 ¼ It is unacceptable to me, 2 ¼ I am not sure whether it is acceptable, 3 ¼ It is acceptable to me). Under-standability was included for it could tell whether English with Chinese features causes understanding problems for its ownspeakers. To make sure nothing else in the examples interrupting participants' judgment except those features investigated,Chinese explanations were added to some difficult words that appeared in the questionnaire and necessary backgroundinformation was also given before the presentation of some conversation examples in the questionnaire.

To investigate reasons underpinning participants' attitudinal responses, 31 students (15males,16 females) and 33 teachers(7 males, 26 females) from the 1782 participants (5e6 teachers and students from each university) were further asked to givereasons to justify their choices to each questionnaire itemwhile they were doing the item-response survey. At the end of thesurvey, they were also required to comment on an open-ended question probing their opinions about having the English usedin the questionnaire in their college English classrooms.

For data analysis, the numerical data recording participants' choices for the 18 questionnaire items were analysedwith thestatistical data processing software of SPSS18.0. Both mean scores and the percentages information about different choices inthe understandability and acceptability dimensions were calculated first by items, and then by categories (i.e. phonology,lexis, syntax and discourse pragmatics) for teachers and students respectively. Considering the disproportion of students(1589) and teachers (193) in the survey, data about the overall acceptability of CE were arrived by averaging the two sets ofscores from teachers and students respectively instead of treating all participants as an indiscriminate whole. To furtheridentify any attitudinal differences of participants to the four categories of CE features and the differences of participants'attitudes to individual items within each category (within subjects differences), repeated measures ANOVAwere conducted.To examine the attitudinal differences between teachers and students (between subjects differences), independent samples t-tests were conducted. For the qualitative data exploring reasons for participants' attitudinal responses to CE, the stepsproposed by Miles and Huberman (1994) for qualitative data analysis were employed to first outline an initial analytic cat-egories, then develop and refine them into a coherent scheme and apply it to analyse all the data, and finally read andcompare the coding results reiteratively to generate any connections within/across categories. To be specific, I first analysedall the reasons given by the participants in justifying their positive/ambiguous/negative evaluations of CE item by item, andparticipant by participant. Then, I grouped all the reasons given by the participants into categories according to the teachers/students divide and the positive/ambiguous/negative attitudes divide respectively. Last, I read and compared the categoriesrepeatedly to find out the underlying criteria participants adopted in their attitudinal responses to CE and the similarities ordifferences between teachers and students in their judging of CE.

4. Findings

4.1. The participants' attitudinal responses to CE features in categories

The quantitative data in the study revealed low acceptability of CE to university teachers and students even though it wasunderstandable to most of them. As Table 1 demonstrates, an overwhelming majority of participants (exceeding 96%) had no

Table 1The acceptability of the four categories of CE features to university teachers and students.

Category Understandability Acceptability M SD F dfs p

1 (%) 2 (%) 1 (%) 2 (%) 3 (%)

Phonology 2.2 97.8 47.5 17.8 34.7 1.87 .80 56.16 2.01/3369.77 .000**Lexis 3.9 96.1 47.7 22.2 30.2 1.83 .45Syntax 2.6 97.4 50.2 17.5 32.3 1.82 .51Discourse pragmatics 3.5 96.5 50.3 23.8 25.9 1.76 .53Total 3.0 97.0 48.9 20.3 30.8 1.82 .41

Notes: 1, 2 in Understandability refer to Don't understand& Understand respectively; 1, 2, 3 in Acceptability refer to Reject, Uncertain, and Accept respectively;**p < .001.

W. Wang / System 53 (2015) 60e7264

problem in understanding expressions with CE features, but only 30.8% chose to accept them. Nearly half of the participants(48.9%) regarded them as unacceptable and another 20.3% were ambivalent towards them.

A repeated measures ANOVA, with Greenhouse-Geisser correction, was conducted to assess possible differences betweenthe average ratings of the four CE categories. Statistical results indicated that participants did rate the four categories of CEfeatures differently, F(2.01, 3369.77) ¼ 56.16, p ¼ .000 < .001, eta2 ¼ .09. The percentages, means and standard deviations forthe acceptability of CE are presented in Table 1 in order from the most acceptable to the least. The statistics suggested thatparticipants rated the phonological features of CE more highly than features in other categories. Polynomial contrastsindicated, in support of this, there was a significant linear trend, F(1, 1678) ¼ 133.86, p ¼ .000 < .001, eta2 ¼ .07. In brief,participants' attitudes towards CE varied across CE categories. Comparatively speaking, CE accents enjoyed higher accept-ability (34.7%) than other features at the lexical (30.2%), syntactic (32.3%), and discourse pragmatic levels (25.9%). What'smore, understandability may not be a determining factor in the evaluation of CE, given a large proportion of participantsunderstood CE expressions in the questionnaire but still chose to reject them.

Table 2 demonstrates that participants not only evaluated the four CE categories differently, but also features within eachcategory. Results of repeated measures ANOVA with Greenhouse-Geisser correction indicated that the differences amongparticipants' responses to individual CE features reached statistical significance in all three CE categories, that is, the lexicalcategory F(4.72, 8294.48) ¼ 380.22, p ¼ .000 < .001, eta2 ¼ .46, the syntactic category, F(4.81, 8448.36) ¼ 299.49,p ¼ .000 < .001, eta2 ¼ .45, and the discourse pragmatic category F(3.87, 6818.39) ¼ 169.12, p ¼ .000 < .001, eta2 ¼ .29.

Generally speaking, high percentages of participants rejected most CE features in syntactical and discourse pragmaticcategories, but their attitudes towards features in the lexical category vacillated. In the survey, three features of CE lexes wereexamined with six questionnaire items. The first feature was the transliteration of Chinese pinyin into English. Two Chinesetransliterations were examineddone was Putonghua (Item 2), the other was Ganbei (Item 3). The two were selected becausethey represented two different developing stages of Chinese transliterations. According to Xu (2010), Putonghua has been

Table 2The acceptability of individual CE features in categories.

Item N Reject Uncertain Accept M SD Fa dfsa p

1 (%) 2 (%) 3 (%)

I2 1778 29.0 17.4 53.6 2.25 .88 380.22 4.72/8294.48 .000**I3 1776 57.0 24.5 18.6 1.62 .77I4 1778 31.8 28.8 39.5 2.08 .84I5 1780 68.5 20.0 11.6 1.43 .69I6 1778 39.2 18.3 42.6 1.55 .74I7 1774 60.1 24.6 15.3 2.03 .85

I8 1775 63.5 17.5 19.1 1.56 .79 299.49 4.81/8448.36 .000**I9 1778 71.2 16.9 12.0 1.41 .69I10 1778 51.3 14.1 34.8 1.84 .91I11 1777 44.7 21.6 33.8 1.89 .88I12 1775 27.0 18.9 54.2 2.27 .86I13 1777 43.3 16.2 40.6 1.97 .90

I14 1777 54.7 16.0 29.4 1.75 .87 169.12 3.87/6818.39 .000**I15 1776 57.1 23.5 19.5 1.62 .79I16 1776 32.4 28.4 39.2 2.07 .84I17 1772 42.1 27.8 30.2 1.88 .84I18 1772 64.6 23.8 11.6 1.47 .69

Notes:**p < .001.

a The F and dfs were adjusted according to the Greenhouse-Geisser correction.

W. Wang / System 53 (2015) 60e72 65

comparatively established in English for some time and has gained popularity in literature. Besides, the concept or the thing itrefers to has no corresponding substitute in native English. Ganbei is still in the developing stage and has an Englishequivalent Cheers. The statistical results demonstrated that more than half of the respondents (53.6%) accepted the estab-lished transliteration of Chinese word Putonghua, but only 18.6% accepted the developing one Ganbei. A paired-samples t testindicated that the differences between the two ratings (Item 2, M ¼ 2.24, SD ¼ .86; Item 3, M ¼ 1.53, SD ¼ .74) were sta-tistically significant, t(1772)¼ 28.97, p¼ .000 < .001, d¼ .89. The second lexical feature included was Chinese loan translationor calque, that is, literal translation of Chinese concepts or idioms into English. One China-specific term (one country, twosystems in Item 4) and one Chinese-culture-bounded idiom (wear the same pair of trousers and breathe through one nostril inItem 5) were examined in the questionnaire. For these two items, the paired-samples t test indicated that participants weremore open to English expressions specific to Chinese concepts or ideas (Item 4, M ¼ 2.09, SD ¼ .83) than the literarilytranslated Chinese idioms or metaphors (Item 5, M ¼ 1.47, SD ¼ .71), t(1776) ¼ 26.93, p ¼ .000 < .001, d ¼ .80. The ratings inTable 2 reveal that 39.5% of the participants accepted Item 4, while only 11.6% accepted Item 5. The third group of ChineseEnglish lexes was English words taking on semantic shifts in the process of nativisation in China, for example, save one's face inItem 6 and back door in Item 7. In the study, 42.6% participants accepted the use of save one's face, but only 15.3% accepted backdoor and the paired-samples t test indicated that the differences between the two (Item 6,M¼ 1.53, SD¼ .74; Item 7,M¼ 1.79,SD ¼ .87) were statistically significant, t(1770) ¼ 12.14, p ¼ .000 < .001, d ¼ .32. The findings seem to suggest that as adeveloping variety, the acceptability of CE needs to be considered case by case. Some usages of CE are more established andhave the potential to be incorporated in classrooms, but others are still under development and not ready to be taught. It maynot be easy to generalise.

4.2. Teachers' and students' attitudes towards CE features in comparison

Table 3 lists the respective ratings of teachers and students to the four categories of CE features in the survey. On thewhole,both the mean scores of teacher participants' ratings and the percentages of teacher participants accepting CE were higherthan those of student participants'. Independent t tests indicated that these two groups differed significantly in their eval-uation of CE in general, t(1679) ¼ 7.81, p ¼ .000 < .001, d ¼ .58, and the differences were particularly salient at the phono-logical level, t(227.5) ¼ 12.89, p ¼ .000 < .001, d ¼ 1.04; the lexical level, t(1731) ¼ 3.92, p ¼ .000 < .001, d ¼ .29, and thesyntactic level, t(1751) ¼ 2.22, p ¼ .000 < .001, d ¼ .18. Differences at the pragmatic level were also found but they did notreach statistical significance, t(1751) ¼ 2.22, p ¼ .06 > .05, d ¼ .18.

The descriptive data further demonstrated that teachers (M ¼ 2.28, SD ¼ .86) evaluated CE phonology much morepositively than students (M¼ 1.46, SD¼ .74). This finding is slightly different from that in the previous study of He and Zhang(2010). Both studies found that CE accents were more acceptable than other features of CE, but their study reported higherstudent acceptability of Chinese accents than teachers (58.2% vs 43.4%). This study, on the contrary, revealed higher per-centages of teachers accepting CE accents than those of students (54.9% vs 14.5%). Besides, this study also found that teacherparticipants seemed to be more positive about CE features at the lexical level than their student counterparts. 33.9% teachersin the survey accepted CE features in the lexical category, but the accepting rate of students was only 22.6%. However, studentparticipants were found to be less resistant to CE features at the syntactic and discourse pragmatic levels than teacherparticipants. 48.4% students rejected CE features at the syntactic level, while the percentages of teachers rejecting CE syntaxreached 51.9%. For CE features at the pragmatic level, 45.6% students rejected and again the percentages of teachers rejectingthem were 54.7%, higher than those of students.

This pattern was largely verified by teachers' and students' respective ratings to almost all CE features examined in thesurvey (Table 4).

Table 3Teachers' and students' attitudes towards CE features in comparison.

Category Reject Uncertain Accept M SD t df p

1 (%) 2 (%) 3 (%)

Phonology Ts 26.6 18.8 54.7 2.28 .86 12.89a 227.5a .000**Ss 68.6 16.9 14.5 1.46 .74

Lexis Ts 45.1 21.0 33.9 1.87 .44 3.92 1731 .000**Ss 50.1 23.5 26.5 1.74 .45

Syntax Ts 51.9 12.9 35.3 1.88 .50 2.22 1751 .027*Ss 48.4 22.2 29.5 1.79 .49

Discourse pragmatics Ts 54.7 18.6 26.7 1.72 .55 �1.90 1759 .058Ss 45.6 29.2 25.2 1.80 .51

Total Ts 44.6 17.8 37.6 1.94 .43 7.81 1676 .000**Ss 53.2 22.9 23.9 1.70 .40

Notes:Ts ¼ Teacher participants, Ss ¼ Student participants.*p < .05, **p < .001.

a The t and df were adjusted because variances were not equal.

Table 4Comparisons of teachers' and students' attitudes towards individual CE features.

Category Reject Uncertain Accept M SD t df p

1 (%) 2 (%) 3 (%)

I1 Ts 26.6 18.8 54.7 2.29 .86 12.89a 227.5a .000**Ss 68.6 16.9 14.5 1.46 .74

I2 Ts 30.4 14.1 55.5 2.06 .90 6.42a 231.5a .000**Ss 27.6 20.7 51.7 1.62 .83

I3 Ts 50.5 26.6 22.9 2.23 .89 1.98 1776 .048*Ss 63.4 22.3 14.3 2.10 .88

I4 Ts 33.3 27.1 39.6 1.69 .82 1.31 1774 .191Ss 30.3 30.4 39.3 1.61 .80

I5 Ts 72.9 16.7 10.4 1.97 .85 �.34 1776 .735Ss 64.0 23.2 12.8 2.00 .83

I6 Ts 25.0 15.6 59.4 1.76 .87 2.94a 227.9a .004*Ss 53.3 21.0 25.7 1.57 .76

I7 Ts 58.3 26.0 15.6 1.52 .74 .185 1775 .854Ss 61.9 23.2 14.9 1.51 .73

I8 Ts 70.3 10.9 18.8 1.98 .93 4.20a 229.6a .000**Ss 56.6 24.0 19.4 1.69 .83

I9 Ts 71.4 16.1 12.5 1.51 .80 �1.11 1772 .266Ss 70.9 17.7 11.4 1.58 .77

I10 Ts 56.3 8.9 34.9 1.47 .77 �1.37 1775 .172Ss 46.2 19.2 34.6 1.55 .78

I11 Ts 46.9 14.1 39.1 1.91 .94 .71a 228.76a .481Ss 42.5 29.1 28.5 1.86 .83

I12 Ts 29.7 15.1 55.2 2.26 .89 �.52 1773 .601Ss 24.3 22.6 53.1 2.29 .83

I13 Ts 37.0 12.0 51.0 2.14 .93 4.74a 233.5a .000**Ss 49.6 20.3 30.2 1.81 .87

I14 Ts 62.0 13.0 25.0 1.63 .86 �3.46 1775 .001**Ss 47.3 18.9 33.8 1.87 .89

I15 Ts 60.9 19.3 19.8 1.59 .80 �1.16 1774 .246Ss 53.3 27.7 19.1 1.66 .78

I16 Ts 36.1 23.0 40.8 2.05 .88 �.62a 230.58a .537Ss 28.7 33.8 37.5 2.09 .81

I17 Ts 42.2 21.4 36.5 1.94 .89 1.85a 229.45a .065Ss 42.0 34.1 23.9 1.82 .79

I18 Ts 72.4 16.1 11.5 1.39 .69 �3.00 1770 .003*Ss 56.8 31.5 11.7 1.55 .69

Notes:Ts ¼ Teacher participants, Ss ¼ Student participants.*p < .05, **p < .001.

a The t and df were adjusted because variances were not equal.

W. Wang / System 53 (2015) 60e7266

The Table revealed that, to most items in the phonological and lexical categories (six to seven), the percentages of teacherswho accepted these expressions were higher than those of students. And the differences were particularly salient in theirattitudes towards CE pronunciation (Item 1), t(227.5) ¼ 12.89, p ¼ .000 < .001, d ¼ 1.04, Putonghua (Item 2, a CE expressiontranslated literarily from Chinese), t(231.5)¼ 6.42, p¼ .000 < .001, d¼ .51, Ganbei (Item 3, a transliterated CE expressionwithan English equivalent), t(1776) ¼ 1.98, p ¼ .048 < .05, d ¼ .51, and save one's face (Item 6, an English expression with meaningexpansion in the Chinese context), t(231.5) ¼ 6.42, p ¼ .004 < .05, d ¼ .51. On the contrary, to most CE expressions in thesyntactic and discourse pragmatic categories, the percentages of teachers who rejected them were higher than those ofstudents. And the differences were particularly significant in Item 8, t(229.6) ¼ 4.20, p ¼ .000 < .001, d ¼ .33, Item 14,t(1775) ¼ 3.46, p ¼ .001 < .001, d ¼ .26, and Item 18, t(1770) ¼ 3.00, p ¼ .003 < .05, d ¼ .23. Item 8 explored participants'attitudes towards using yes in a negative reply such as the yes in A: You mean your hometown is not so crowded? B: Yes, not socrowded. 70.3% teachers rejected such uses, but only 56.6% students rejected them. Item 14 examined participants' percep-tions of CE expressions following Chinese pragmatics, such as addressing people by their professional titles in Teacher Zhang.62% teachers rejected it whereas the percentages of students rejecting this expression were 56.6%. Item 18 investigated CEexpressions juxtaposed in the Chinese contextual schema, such as introducing the arrangement of Chinese university stu-dents' dormitory in a way like Item 18. 72.4% teachers and 56.8% students rejected it. The descriptive data in Table 4 alsodemonstrated that students had persistently higher degrees of uncertainties than their teacher counterparts towards most CEfeatures, particularly features at syntactic and discourse pragmatic levels. Findings of such seemed to reveal that the twocohorts held different views towards CE, though both were in educational settings. Teachers were comparatively morepositive about CE pronunciations and lexes, whereas students were less resistant to English constructed in CE syntax anddiscourse pragmatics. To incorporate CE features into classrooms, the attitudinal differences between teachers and studentsneed to be considered.

W. Wang / System 53 (2015) 60e72 67

4.3. Reasons underpinning the participants' attitudinal responses to CE

Instead of having a laundry list of reasons given by participants for their attitudinal responses to CE, report in this sectionfocused on the underlying criteria towhich participants referred in their negative/positive/ambiguous evaluations of CE. Afterrepeated reading and comparison of all explanations given by participants, four criteria were identified to work interactivelyin their evaluations: the correctness of CE against native English standards, the Chinglish overtone, the communicativeness ofCE, and the role of CE for identity making. For thosewho rated CE negatively (48.9% of thewhole population investigated), thefirst two criteria seemed to outweigh the last two in their evaluations. But for those who rated CE positively (30.8% of thewhole population investigated), the last two seemed to be their priorities, while the first two were comparatively uncon-cerned. As to those showing ambiguity in their evaluations (20.3% of the whole population investigated), it seemed that theywere in constant tensions in aligning with the two sets of incompatible criteria.

4.3.1. Criteria foregrounded in the participants' negative evaluation of CETo be specific, correctness was found to be the top consideration of participants who evaluated CE negatively. Furthermore,

in their judgments of the correctness of these CE expressions, native-speaker norms were usually the de facto yardsticks. Forinstance, seven participants explicitly pointed out that Ganbei was Chinese and should be replaced with Cheers, its Englishcounterpart. They argued that:

1 I3-TBesides

In Chinese we say Ganbei, but in English we should say Cheers. (I3-T201)

In this sentence, English intertwines with Chinese. For the use of Chinese Ganbei [in English], the speaker takes toomuch for granted. (I3-S12)

Both excerpts expressed a strong belief that they should observe English ‘standards’ while speaking English. These par-ticipants seemed to believe that Ganbei was Chinese whereas Cheers was English. It was undesirable to infuse Chinese pinyininto English expressions. Another sixty-eight instances were found refusing the syntactic features of CE in Items 9, 10 and 11for their violation of ‘standard’ English grammar.

Grammar mistake, two subjects (I9-S4)

It is not allowed to put two clauses in one sentence without a conjunction. Unacceptable to natives. (I11-T29)

As suggested in these excerpts, to judge the acceptance of CE expressions, what came first for the participants was theiragreement with the grammatical rules of native speaker English. Communicativeness seemed not to be an issue of concern.Besides seeking linguistic norms from inner circle English, the participants were also observed to resort to Anglo-Americancultures and value systems when evaluating the appropriateness of English usages. For instance, the participants explainedwhy it was improper to address people Teacher Zhang or respond to others' compliments in a self-depreciating manner in thefollowing:

In English speaking countries, Teacher Zhang should be Ms or Mr. Zhang (I14-S16)

It is Chinese tradition to deliberately grade down oneself in response to others' compliments. But in English, we shouldobserve English-speaking countries' customs and express thanks for others' compliments. (I15-T31)

These comments on Teacher Zhang again revealed a native speaker oriented view of English. It seemed to say that whethercertainpractices of speaking Englishwere proper needed to be gauged against native English cultures and conventions, but notthe appropriateness of them to the actual communicating situations. Furthermore, these comments also implied that theparticipants did not sense theneed to present the Chinese reality in English. It seemed that the participants knew Teacher Zhangwas a conventionalway to address teachers in Chinese, but to speak English, oneneeded todo as those English speakers did andforgot their native culture of being a Chinese. Given a large proportion of the participants in the survey evaluated CE ex-pressions negatively, one may infer that native speaker English was still the dominant ideology in the educational settings ofChina, and was also the determining factor against which teachers and students' English practices were judged.

The second frequently referred criterion in the participants' negative evaluation of CE was whether the expressions tookon any Chinglish overtones. In the survey, expressions with CE features had often been taken as Chinglish and rejected. Forexample, five participants simply chose 1 while rating the acceptability of CE expressions and wrote Chinglish as their ex-planations for all items in the questionnaire. Another large proportion of participants (nearly half of them) were found toimplicitly relate the expressions in the survey with Chinglish or Chinese English. To them, expressions taking on Chinesecharacteristics were deviations from native speaker English, and thus, were Chinglish, were half-products generated byEnglish learners for their poor control of the Chinese interferences. The two criteria seemed to go hand in hand. They did notseem to recognise the potentiality of CE in communicating local cultures and values.

20 refers to the 20th teacher participant's response to Item 3 in the survey. All tags after quotations are numbered in the same way in this paper., all quotations originally in Chinese were translated into English when presented in this paper.

W. Wang / System 53 (2015) 60e7268

These are simple Chinglish. The speaker uses English in a totally Chinese way without taking into account Englishculture and conventions. If we speak English this way, we could only get ourselves despised. When we are learning alanguage, we should also know the culture of the country which speaks that language. Only in this way, canwe learn alanguage well. (I19-S19)

The sentences in this questionnaire are full of grammatical mistakes and expressions seriously interfered by Chineseculture or Chinese language. It is natural for English learners to produce such flawed English. But we teachers shouldhelp students correct them and teach them some English learning skills for them to improve their English. (I19-T7)

It seemed that the Chinglish stigma had been deeply rooted in the minds of these participants. Just as revealed in I19-S19, ‘speaking in a Chinese way’ was simply equated as Chinglish, That was serious for him, because just as he argued, “If wespeak English this way, we could only get ourselves despised”. We may see from his statement that the essence for theirconcern of deviating from native English norms or cultures was nothing else but the serious social consequence accom-panied with Chinglishdgetting insulted in the society. Due to such serious impacts of speaking Chinglish, they would avoidexpressions carrying any Chinglish overtone. Therefore, the participants in the survey would rather take English withChinese features as learner language under improvement instead of as powerful means for self-expression. If this was thecase, it might be desirable to move them out of the Chinglish mindset before any Chinese English feature was introduced inclassrooms.

Communicativeness was also a criterion leading to participants' negative evaluation of CE, but it was less frequentlymentioned as the previous two. Even in most cases where communicativeness was concerned, native speaker English wasoften indirectly indexed as shown in the following examples. These were comments on the literal translation of a Chineseidiom wear the same pair of trousers and breathe through one nostril in Item 5:

Oh, my gosh! The expression is too Chinese. How could you expect a foreigner to understand this?! If the Americanscould, you don't need to speak English. Go direct with our Chinese. They are sure to understand! (I5-S19)

It is fine to speak like this between the Chinese. But to the westerners, I guess they will not understand it. (I5-T20)

In these comments, understandability of CE constituted participants' major concern. But with further reading, one mayfind that the projected communities of English users in these comments were primarily native speakers. Evenwith necessarybackground information reminding them the communicating situations in the questionnaire survey, the presumed in-teractants in these conversations were still ‘foreigners’, or more explicitly ‘westerners’, or even narrower, ‘the Americans’din aword, native speakers. They rejected the literal translation of Chinese idioms in English for the concern that the natives mightnot understand them if these English expressions were too Chinese-specific. By implication, the above comments seemed toindicate that Chinese characteristics were only supposed to be appropriate within Chinese communities. There was no needto communicate the Chineseness to the global audience.

4.3.2. Criteria foregrounded in the participants' positive evaluation of CEContrary to participants who took negative views of CE, those who perceived CE positively seemed to apply different

criteria in evaluating CE. They value the capability of CE in fulfilling communication purposes and the role of CE for identitymaking more than following native speaker English correctness and avoiding any Chinglish overtone while speaking English.As can be seen from the following explanations:

Although it is not standard English, it does not affect our communication. (I1-T16)

To be understood is enough. In daily communication, it is unnecessary to focus solely on grammatical rules and nativeexpressions. (I19-S16)

These comments revealed that for participants who were concerned mainly about communication rather than standards,CE was not a problem. In other words, CE would be acceptable if it could fulfil the communicative purpose, but not necessarilyits meeting of native English standards as emphasised by those negative evaluators.

The other criterion foregrounded in the participants' positive evaluation of CE was the role of CE in expressing Chineseidentity. For example, a dozen of the participants explicitly aligned the use of CE with their Chinese culture and identity,contending that it was exactly theway the Chinese speak English. Three even talked about the potential contributions of CE tothe future development of English:

As a Chinese speaker of English, I think it is acceptable. (I1-T18)

We often say in that way. Personally, this expression is very idiomatic. It is quite appropriate to be used to expressChinese culture. (I13-S10)

Back door, though a Chinese expression, should be acceptable as a good metaphor to westerners, just like behind thetable. (I7-S15)

English culture and English language are tolerant, so you can use that language the way you like. You may enrich thatlanguage by using it in a Chinese way. (I19-T5)

W. Wang / System 53 (2015) 60e72 69

These comments pointed to the needs to express Chinese culture and identity in CE, as highlighted by I7-S15 in his choiceof the modal verb should. At the same time, these comments also flagged the issue of norms in speaking English. It is normallyagreed that English in outer circle countries is norm-independent or endonormative. It has increasingly derived the norms ofcorrectness and appropriacy from locally-relevant practices. But in expanding circle countries, English is still norm-dependent or exonormative, that is, relying on native/standard English for correctness, for it has not been widely used inthese local contexts yet (Kachru, 1985; Matsuda & Friedrich, 2012). The comments here seemed to indicate that even inexpanding circle countries like China, people also began to break away from exonormative models of English and starteddeveloping norms of their own. But this only happened when English was connected to the expression of local cultures andidentities.

4.3.3. Conflicting criteria in participants' ambivalent evaluation of CEIt is also noteworthy that a certain number of participants (20.3%) expressed their uncertainties toward CE. On the one

hand, they felt the English expressions in the questionnaire were close to their daily experiences. On the other hand, theydoubted whether such expressions were legitimate. Their ambiguous attitudes were reflected in the following excerpts:

Almost every student of mine speaks English this way and I am quite used to it. They have never gone abroad. You can'texpect them to speak like native speakers in an all-Chinese environment. (I19-T4)

Right now we hear a lot of young people speak that way. It is a common way to express Chinese culture. There is noother special ways. (I5-S24)

It seems that some Chinese English users have recognised the legitimacy and rightfulness of CE for local needs. However,they were worried about any negative value judgment laid on the Chinese way of using English in society. One studentparticipant expressed:

If all of us use English as described in the questionnaire, it will be terrific. Learning English will be much easier for us.But if it is only encouraged in the classroomwhile the outsideworld doesn't accept it, I will not takemy teacher's advice[of using CE]. (I19-S28)

Such responses reflected the dilemma that the participants experienced in learning English. They wished to use theEnglish that was close to them but also felt the need to attain approval from the society. Their choice of preferred English wasrestrained by constant tension between speaking English with local characteristics and meeting the socially recognisedstandards. Participants' unwillingness to accept CE might not arise from their ignorance of the power of CE for localexpression, but because the dominant ideology at work in society was so strong that they could not penetrate it to appropriateEnglish at free will.

4.3.4. Teachers/students divide in CE evaluationsAs to variations between teachers' and students' attitudes towards CE, there were different explanations as specific CE

features involved. For example, one teacher participant gave the following comment on CE pronunciation:

There is no language police. We should not prescribe what way can be accepted and what way cannot. In addition, Ihave chattedwith some people from India, Japan, Hong Kong and Korea. I knowhow they speak English. It is no big dealto have a Chinese accent. (I19-T5)

This participant's exposure to other non-native English varieties seemed to influence his attitude towards CE pronunci-ation. Another teacher accepted Ganbei, because he once heard a foreigner using it:

Ganbei is as acceptable as Zaijian for the foreigners. I once heard them use it. (I3-T14)

However, quite a few students were less tolerant towards CE pronunciation, pointing out that it did not sound British orAmerican. Some other students considered one country, two systems unacceptable because they assumed there was not suchan expression in English:

One country, two systems is unique to China. Foreign countries don't have. We need to explain to them but not totranslate to English literally. (I4-S29)

Another example in point is that many teachers expressed their refusal of English expressions situated in Chinese-featureddiscourse because of their poor cohesion and coherence from the ‘native’ English perspective, but students tended to acceptthem because these expressions were closer to their everyday use of English.

How does the first sentence relate to the second one? And the sentence nextdpeople in America and China can talk witheach other directly, is it because of the invention of telephones? The last sentence, what does it stand for? This para-graph is in poor coherence and cohesion. (I16-T29)

This English might not be as good as American or British English, but it is close to our life in China. We talk in the sameway in our oral English class. Our dorm is organised like this. For those who haven't seen it, no matter how hard I try,they may have no idea of it. For those who know, they will understand the moment they hear what I say. (I18-S21)

W. Wang / System 53 (2015) 60e7270

Generally speaking, teachers might have more exposures to the actual use of English worldwide than students and morechances to examine the communicativeness of certain CE features like Chinese idioms and metaphors in internationalcommunications, and thus were more open to such usages, whereas students might have more personal experiences of usingcertain specific CE expressions, and thus being more tolerant to them.

5. Discussions

By examining participants' attitudinal responses towards specific CE features, this study has built up on previous ones toelaborate specific CE features acceptable to university teachers and students and the reasons underlying their attitudinalresponses. The findings of this inquiry largely conform to previous studies that CE has not been pervasively accepted yet(e.g., Kirkpatrick & Xu, 2002; He & Li, 2009; He & Zhang, 2010; Hu, 2004, 2005; Yang & Zhang, 2015), but some features dogain high recognitions and have the potentiality to satisfy non-inner-circle English users' desire for local expression. Forinstance, the study revealed that the participants were comparatively open to CE accents, but resistant to English mixedwith Chinese discourse pragmatics and the usages of English violating grammar rules of native speaker English. The studyalso has contradictory findings in comparison with He and Zhang (2010) regarding teachers' and students' attitudes towardsCE pronunciation. They found “in terms of pronunciation, students displayed more willingness to accept CE [China English]than their teachers” (He & Zhang, 2010, p. 784). But this study arrived at opposite conclusionsdthe percentages of teachers(54.9%) accepting CE pronunciation was far higher than those of students (14.5%). The reasons given by the focal partici-pants in the survey revealed that it might be because teachers had more experiences of getting contacts with differentEnglish accents internationally and thus were more tolerant to phonological variations, whereas students usually projectedspeaking English with native speakers and thus were aspiring to achieve native-like accents. Our different findings may beattributed to the use of different research methods. This study used an authentic CE speakers' speech to elicit participants'responses towards CE pronunciation, whereas they asked their participants to choose one speaking model they would liketo align with according to the description of two speakers, one was described as speaking native-like English and the otheras speaking English with local accents. Without real CE examples, participants might have different local English images inmind while giving out their ratings in the survey.

Findings of this study have some important implications for the incorporation of local English varieties in Englishteaching and learning practices. In agreement with previous studies, the high degree of rejection of CE features in all fourcategories renders the adoption of local English varieties as pedagogical models unwise in the context of China, but theovert recognition of some CE features among a certain amount of participants in the study lends possibility for the inclusionof these features for pedagogical purposes (He & Li, 2009; He & Zhang, 2010; Hu, 2004, 2005). To be specific, the highacceptability of CE pronunciation indicates that teachers should not overemphasise native-like standards in speaking En-glish, but allow students to keep some Chinese accents if they do not impede communication. For CE at the lexical level,some well-codified and successfully promoted Chinese transliterations such as Putonghua, translations of China-specificconcepts such as one country, two systems, and nativised English words with semantic shift such as save one's face mightbe possible to be introduced into classrooms. At the syntactic level, it is better to follow the grammatical rules of ‘standard’English as most teacher and student participants in the investigation aspired. At the discourse pragmatic level, teachersneed to consider carefully how to facilitate students to express local cultures and ideas and at the same time maintaininginternational communicativeness because both of the two aspects were serious concerns for classroom practitioners asidentified in the investigation. Admittedly, this study only investigated some often discussed CE features under codificationin the literature. More studies are needed to gain a fuller understanding of the acceptability of CE features among languageteachers and learners.

The participants' explanations for their refusal of CE features further indicate that native speakerism and Chinglish remainthe predominating ideologies in the educational context of China. Native speakerism and Chinglish are actually the two sidesof a same coin. Because native speaker English is taken as standards and norms, English deviating from these norms is treatedas unacceptable Chinglish. We may derive from the widespread discourse of Chinglish that university teachers and studentsin China are still at the lower end of the Chinglish/China English continuum and do not recognise the legitimacy of CE inexpressing local needs and culture. If this is the case, I argue that the simple promotion of some CE features in pedagogy maynot be enough. More importantly, it has become necessary to raise learners and teachers' awareness of the sociolinguisticrealities of English and move them out of the native English/Chinglish mindset. Only with the necessary shift in mind, will itbe possible for teachers and learners to appreciate the value of locally relevant English. However, I understand that languageideologies are created in social, cultural and political contexts. If standard English and Chinglish are still the dominant ide-ologies in society, it is difficult for individuals to pursue self-liberation. It is the change of the social discourse that finallyreshapes individual's language attitudes. As Sharifian (2009) argued, “[I]ssues such as identity, ideology and power aredirectly relevant to and do have a determining role on the content as well as the approach in ELT” (p. 11).

The differences in teachers and students' attitudes towards CE as observed in the inquiry also imply that teachers need topay attention to such attitudinal gaps when promoting usages of local English varieties in classrooms. For instance, in thisstudy it is found that student participants were willing to use English with Chinese pragmatic and discourse features becausesuch English was closer to their lives. For teachers in this situation, it might be inappropriate to present a restricted set of pre-described CE features to learners. Instead, they would better tailor English education according to learners' situationalcommunication needs and empower learners to appropriate possible local English resources for self-expression.

W. Wang / System 53 (2015) 60e72 71

6. Conclusion

The study on the attitudes of university teachers and students towards CE reveal that as a developing variety, it has not yetbeen widely recognised by its speakers. Most of the features in pronunciation, lexis, syntax, and discourse pragmatics dis-cussed in literature have received low acceptability in the investigation. However, the study did find some features enjoyingcomparatively high recognition by both teacher and student participants, which indicate the possibility to be included inEnglish language teaching. Nevertheless, to teach CE to enhance its role in expressing local cultures and ideas, more workneeds to be done to change Chinese learners' native speaker mindset besides the incorporation of some well-accepted CEfeatures in pedagogy. Further research is needed tomonitor the evolving attitudes and perceptions of learners and teachers incontexts like China to see whether and how local varieties are being accepted as desirable pedagogical goals before inte-grating them into actual learning and teaching practices.

Acknowledgements

The work was supported by the Educational Bureau of Hubei, China under the Provincial Social Sciences ResearchGrant [No. 13g049]. The author would like to thank all the teachers and students who had participated in the project andthe two anonymous reviewers for their helpful comments on the earlier versions of this paper. All remaining errors aremy own.

Appendix. China English Features examined in the survey

Phonology1 (Play a recording of a Chinese speaker introducing his family in English) There are four people in my familydmy father, my mother, mybrother and me. My father is a teacher and my mother is a nurse. My brother is threeyears old while I am thirteen. We love each other very much.Lexis2 (Transliteration) He speaks putonghua fluently.3 (Transliteration) For a toast everybody gets up, raises their glasses and touches the others' glasses, saying Ganbei!4 (Loan translation) The concept of one country, two systemsdpreserving two different political, social and economic systems within one nation hasno precedent.5 (Loan translation) You mean Xiao Lan, Xiao Li and Xiao Liu? Oh, they three wear the same pair of trousers and breathe through one nostril.6 (Semantic transfer) In order to save her face, he decided not to tell her classmates the truth.7 (Semantic transfer) Our manager is his uncle, so he gets this position by using the back door.Syntax8 (Yes-no response) A: You mean your hometown is not so crowded? B: Yes. Not so crowded.9 (Subject pronoun copying) Some of my college classmates they like to dress up very much.10 (Adjacent default tense) When I was very young, I want to be a doctor, but now I want to be a teacher in universities.11 (Multiple-coordinate construction) In the present society, love is not important, everyone cares about money.12 (Modifying-modified sequencing) Because many farmers lack adequate knowledge and experience to distinguish counterfeit goods they aregenerally more vulnerable.13 (Top heavy lion style structure) For me to get up before 6 o'clock in the morning is impossible.Discourse pragmatics14 (Chinese pragmatic notions) Good morning, Teacher Zhang! (Greeting)15 (Chinese pragmatic notions) A: You speak very good English! B: No, no, my English is not good at all.16 (Inductive pattern of reasoning)A: Is the invention of telephone a convenience or a nuisance for us?B: With the development of science and technology, telephone can be found everywhere, especially in big cities. There are many people talkingwith their friends in USA directly. We can know what is happening at one place thousands of miles away immediately. Can you imagine how longit would take to send a letter to USA? At least two weeks. So, it brings much convenience to us, particularly in communication.

17 (Discourse in line with Chinese culture and value systems) (Jinli was a mother who left her daughter Dandan and husband Chigan in China andwent to New York alone. A few years later, she came back to see her daughter who was living with her husband's parents at that moment. Belowis a conversation between her and her mother-in-law.)“When… when will she come home?” Jinli asked.“This is her home,” said Chigan's mother.“Please, let me have a look at her. Tears were gathering in her eyes, but she tried suppressing them.“No. She doesn't want to be disturbed by you.”“Mother, forgive me just this once, please!”“Don't call me that. You're not my daughter-in-law anymore.”The door was shut.

18 (Discourse in line with Chinese situational schema)Visitor: How many people are there in your dormitory?Student: A big room has three /ru:/…eh..a flat has three rooms. In the flat, there are ten people. In the smaller room there are three. In my roomthere are four.

W. Wang / System 53 (2015) 60e7272

References

Alsagoff, L. (2010). English in Singapore: culture, capital and identity in linguistic variation. World Englishes, 29(3), 336e348.Bolton, K. (2003). Chinese Englishes: A sociolinguistic history. Cambridge: Cambridge University Press.Bolton, K. (2012). World Englishes and Asian Englishes: a survey of the field. In A. Kirkpatrick, & R. Sussex (Eds.), English as an international language in Asia:

Implications for language education (pp. 13e26). New York: Springer.Bruthiaux, P. (2010). World Englishes and the classroom: an EFL perspective. TESOL Quarterly, 44(2), 365e369.Brutt-Griffler, J. (2002). World English: A study of its development. Clevedon: Multilingual Matters.Chan, J. Y. (2013). Contextual variation and Hong Kong English. World Englishes, 32(1), 54e74.Cook, V. (2007). The goals of ELT: Reproducing native-speakers or promoting multicompetence among second language users?. In International handbook of

English language teaching (pp. 237e248) New York: Springer.Deterding, D. (2006). The pronunciation of English by speakers from China. English World-Wide, 27(2), 175e198.Fang, F. (2008). People mountain, people sea: a study of four Chinese English idioms on the Web. English Today, 24(4), 46e50.Gao, X. (2012). The study of English in China as a patriotic enterprise. World Englishes, 31(3), 351e365.Gao, X., Liao, Y., & Li, Y. (2014). Empirical studies on foreign language learning and teaching in China (2008e2011): a review of selected research. Language

Teaching, 47(1), 56e79.Ge, C. (1980). Mantan you han yi ying wenti [Talking about some problems in Chinese-English translation]. Fanyi Tongxun [Translator's Notes], 2, 1e8.Gil, J., & Adamson, B. (2011). The English language in mainland China: a sociolinguistic profile. In A. Feng (Ed.), English language education across Greater

China (pp. 23e45). Buffalo, NY: Multilingual Matters.He, D., & Li, D. C. S. (2009). Language attitudes and linguistic features in the ‘China English’ debate. World Englishes, 28(1), 70e89.He, D., & Zhang, Q. (2010). Native speaker norms and China English: from the perspective of learners and teachers in China. TESOL Quarterly, 44(4), 769e789.Henry, E. S. (2010). Interpretations of ‘Chinglish’: native speakers, language learners and the enregisterment of a stigmatized code. Language in Society,

39(5), 669e688.Hino, N. (2009). The teaching of English as an international language in Japan: an answer to the dilemma of indigenous values and global needs in the

expanding circle. AILA Review, 22(1), 103e119.Holliday, A. (2005). The struggle to teach English as an international language. Oxford: Oxford University Press.Hu, X. Q. (2004). Why China English should stand alongside British, American, and the other ‘World Englishes’. English Today, 20(2), 26e33.Hu, X. Q. (2005). China English, at home and in the world. English Today, 21(3), 27e38.Jenkins, J. (2006). Current perspectives on teaching World Englishes and English as a Lingua Franca. TESOL Quarterly, 40(1), 157e181.Jenkins, J., Cogo, A., & Dewey, M. (2011). Review of developments in research into English as a Lingua Franca. Language Teaching, 44(3), 281e315.Kabel, A. (2009). Native-speakerism, stereotyping and the collusion of applied linguistics. System, 37(1), 12e22.Kachru, B. (1985). Standards, codification and sociolinguistic realism: the English language in the outer circle. In R. Quirk, & H. G. Widdowson (Eds.), English

in the world: Teaching and learning the language and literatures (pp. 11e33). Cambridge: Cambridge University Press.Ke, I.-C., & Cahyani, H. (2014). Learning to become users of English as a Lingua Franca (ELF): how ELF online communication affects Taiwanese learners'

beliefs of English. System, 46, 28e38.Kirkpatrick, A. (2006). Which model of English: native-speaker, nativized or lingua franca? In R. Rubdy, & M. Saraceni (Eds.), English in the world: Global

rules, global roles (pp. 71e83) London: Continuum.Kirkpatrick, A. (2007). World Englishes: Implications for international communication and English language teaching. Cambridge: Cambridge University Press.Kirkpatrick, A., & Xu, Z. (2002). Chinese pragmatic norms and ‘China English’. World Englishes, 21(2), 269e279.Kumaravadivelu, A. (2003). Problematizing cultural stereotypes in TESOL. TESOL Quarterly, 37(4), 709e719.Lam, A. S. (2005). Language education in China: Policy and experience from 1949: Policy and experience from 1949. Hong Kong: Hong Kong University Press.Matsuda, A., & Friedrich, P. (2012). Selecting an instructional variety for an EIL curriculum. In A. Matsuda (Ed.), Principles and practices of teaching English as

an international language (pp. 1e14). Bristol: Multilingual Matters.McKay, S. L. (2010). English as an international language. In N. Hornberger, & S. L. McKay (Eds.), Sociolinguistics and language education (pp. 89e115). Bristol:

Multilingual Matters.McKenzie, R. M. (2008). The role of variety recognition in Japanese university students' attitudes towards English speech varieties. Journal of Multilingual

and Multicultural Development, 29(2), 139e153.Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks: Sage.Morrison, R., & White, M. (2005). Nurturing global listeners: increasing familiarity and appreciation for World Englishes. World Englishes, 24(3), 361e370.Nunan, D. (2003). The impact of English as a global language on educational policies and practices in the Asia-Pacific Region. TESOL Quarterly, 37(4),

589e613.Sasayama, S. (2013). Japanese college students' attitudes towards Japan English and American English. Journal of Multilingual and Multicultural Development,

34(3), 264e278.Seidlhofer, B. (2011). Understanding English as a Lingua Franca. Oxford: Oxford University Press.Sharifian, F. (Ed.). (2009). English as an international language: Perspectives and pedagogical issues. Bristol: Multilingual Matters.Sifakis, N. (2009). Challenges in teaching ELF in the periphery: the Greek context. ELT Journal, 63(3), 230e237.Suzuki, A. (2011). Introducing diversity of English into ELT: student teachers' responses. ELT Journal, 65(2), 145e153.Timmis, I. (2002). Native-speaker norms and international English: a classroom view. ELT Journal, 56(3), 240e249.Waters, A. (2007). Native-speakerism in ELT: plus ça change…? System, 35(3), 281e292.Wei, R., & Su, J. (2012). The statistics of English in China. English Today, 28(3), 10e14.Wei, Y., & Fei, J. (2003). Using English in China. English Today, 19(4), 42e47.Wen, Q. F. (2012). Teaching English as an international language in mainland China. In A. Kirkpatrick, & R. Sussex (Eds.), English as an international language

in Asia: Implications for language education (pp. 79e124). NY: Springer.Xu, Z. (2006). Rectifying ‘Chinese English’. In A. Hashim, & N. Hassan (Eds.), Varieties of English in Southeast Asia and beyond (pp. 283e291). Kuala Lumpur:

University of Malaya Press.Xu, Z. (2010). Chinese English: Features and implications. Hong Kong: Open University of Hong Kong Press.Yang, C. S., & Zhang, L. J. (2015). China English in trouble: evidence from dyadic teacher talk. System, 51, 39e50.Yang, J. (2005). Lexical innovations in China English. World Englishes, 24(4), 425e436.Yoo, I. W. (2014). Nonnative teachers in the expanding circle and the ownership of English. Applied Linguistics, 35(1), 82e86.Young, T. J., & Walsh, S. (2010). Which English? Whose English? An investigation of ‘non-native’ teachers' beliefs about target varieties. Language, Culture

and Curriculum, 23(2), 123e137.Zhang, H. (2002). Bilingual creativity in Chinese English: Ha Jin's in the pond. World Englishes, 21(2), 305e315.