Learners’ views of (non)native speaker status, accent, and identity: an English as an international language perspective

2.1. Native speakers, nonnative speakers, and circles of English

English, regarded as an international lingua franca (ELF), is the most widely taught and learned foreign language in the world. Today, English is perceived as being spoken in three types of context as reflected in Kachru’s (1985) frequently cited concentric circle model. Kachru (1985) argues that nowadays English is spoken in three “circles” in the world. In the Inner Circle which includes English-speaking countries such as the UK, the US, Canada, and Australia, English is spoken and learned as a first language (L1) or as a native language (ENL). The Outer Circle, also called the Extended Circle, includes countries such as India and Malawi where English is spoken as a second language (ESL) and is used for communicational, official, and institutional purposes. Finally, English is taught, learned, and spoken as a foreign language (EFL) in the Expanding Circle, or the Extending Circle, which includes countries like Iran, China, and Japan. Figure 1 below displays the three circles of English as outlined by Kachru (1985). It is noteworthy that a similar figure can be found in also in Kachru (2011, 27).

Figure 1.

The three circles of English (Crystal 2003, 60).

As can be seen from the above figure, while the NNSs of English vastly outnumber its NSs and while, in the context of English as an international language (EIL), most communication occurs among NNSs and thus mutual intelligibility and comprehension should be given the highest priority (see Sifakis and Sougari 2005), native-like accent continues to receive the most attention in language pedagogy on the part of both teachers and learners. Similarly, Moyer (2007) took intelligibility as the departure point to discuss whether accent still assumes central importance in language learning to find out that immigrant learners of English in the US continue to have positive attitudes toward NS accents. Most recently, Carey, Sweeting, and Mannell (2015) argued that pronunciation instruction has long been strongly driven by native-speakerism (see Holliday 2005) and that almost all instructional materials aim at accent reduction, recommending that learners drop their foreign accents while speaking English. In this regard, native-speakerism is defined as “an unjustified and unfair favouritism to native speakers of English at the expense of non-native speakers” (Feyér 2012, 20).

2.2. Attitudes toward accent and identity

A rich body of scholarly literature exists on the intersection of accent and identity. Attitudes toward accent and pronunciation have received considerable research attention throughout the past few decades. Accent has been examined with regard to its relation to the negotiation of identity and participation in academic communities (Morita, 2004), learners’ ethnic group affiliations in ESL context (Gatbonton, Trofimovich, and Magid 2005), social pressure and identity negotiation (Lefkowitz and Hedgcock 2006), learners’ identity and motivation (Feyér 2012), and a wide range of other pertinent concepts and areas of research. This long and rigorous strand of research into accent and pronunciation has been mostly suggestive of learners’ positive attitudes toward NS accent(s) and unfavorable perceptions toward NNS accent (e.g., Cenoz and Lecumberri 1999; El-Dash and Busnardo, 2001; Hartshorn 2013; Kim 2008; Majanen 2008; McGee 2009; Soukup 2011; Walker and Zoghbor 2015). Kim (2008) even found out that ESL students rate foreign-accented speech negatively regardless of whether or not it was intelligible, confirming that intelligibility and foreign accent are two distinct and independent issues. On the other hand, Bresnahan et al. (2012) concluded that intelligibility has positive effects on learners’ attitudes toward NNSs’ accented speech. In addition, most research studies suggest that language learners hold negative attitudes toward nonnative English teachers, preferring native English teachers or nonnative English teachers who are native-accented (Butler 2003; Hartshorn 2013; Timmis 2002). Scales et al. (2006), for instance, examined the perspectives of 37 learners of English and 10 American native speakers on native and nonnative accents of English. The results showed a strong preference for NS accents and a remarkable correlation between the accent that was perceived by the participants to be the easiest to understand and the one that they preferred. On the other hand, the majority of the participants could not perceive the accents that they preferred to have. Scales et al. (2006) interpret this finding as the mismatch between learners’ desires and their true achievement. Negative attitudes toward nonnative-accented speech are prevalent among language teachers in addition to learners. Sifakis and Sougari (2005), for instance, found that Greek EFL teachers hold negative, stereotypic, mostly NS norm-bound, perceptions toward NNSs’ pronunciation. Likewise, Jenkins (2005) found out that nonnative English-speaking teachers (NNESTs) had grave reservations about whether or not to identify themselves as legitimate users of English. Research has revealed that, from learners’ viewpoints, accent must be regarded as a determining criterion in NNS teachers’ employability with NS accent as a positive point for teachers (Alenazi 2012). Researchers have also attempted to uncover the reasons for which learners view NNSs’ accents unfavorably. Baugh (2000) provides evidence of mockery, racism, ridicule, and discrimination against foreign-accented speakers of English who were perceived to have “funny accents.” This finding is further supported by research carried out in Inner Circle contexts such as the US where NSs have been reported to describe NNS varieties of English as “accented,” “broken,” and “little” (see Lindemann 2005). The beauty and attractiveness of NS accent have also been mentioned as a source of learners’ favorable attitudes toward it in different contexts such as Oman (Soukup 2011) and Denmark (Jarvella et al. 2001). Similarly, in a study of the attitudes of Hong Kong learners of English toward different varieties of English, Zhang (2010) found that Hong Kong English accent (HKed) was viewed with solidarity although American English was the most preferred variety. Zhang (2010) regarded such attitudes as expressing “linguistic self-hatred.” Despite the respectable stockpile of research on learners’ attitudes toward and their perceptions of accent and whether and how it relates to learner identity, this issue has not been systematically investigated in the context of Iran as an Expanding Circle context. Put differently, the study aims at giving learners a voice on the issue of accent and its relationship with identity where it is believed that learners’ voice has gone unheard (see Timmis 2002). Recently, researchers developed an interest in the possible connection of identity and pronunciation in a range of contexts. Waniek-Klimczak, Rojczyk, and Porzuczek (2015), for instance, investigated the extent to which Polish learners were willing to display their identities through their foreign accents. In another study motivated by the dearth of research into learners’ beliefs about the qualities of pronunciation teachers, Levis (2017) posited that while students’ beliefs were more inclined toward NEST teachers as better fitting pronunciation classes, their beliefs could be modified by stressing “professionalism as key to effective pronunciation teaching.”

More specifically, the objective of the present study was four-fold in that the study set out to explore the following issues from the viewpoints of Iranian EFL learners: (a) perceptions of and attitudes toward pronunciation and accent, (b) the significance of pronunciation and accent for communication, (c) the factors that impact on pronunciation or accent, and (d) how accent is conceived as related to one’s identity.

3.1. Participants

The participants of the current study consisted of 51 Iranian EFL learners who studied English at the males’ department of a popular private language institute in Ahvaz, Iran. They comprised three classes to which the researcher had access for the purpose of the study and were randomly selected. The participants constituted males only, fell within the age range of 14–26 (14–18: 90%, 19–26: 10%; mean age = 15.9 years) and were of the following ethnic backgrounds: 41 Persians (80%), 9 Arabs (18%), and 1 Turk (2%). With regard to their proficiency level, the participants comprised pre-intermediate (43%) and intermediate (57%) learners and their English language learning experience ranged from 2 to 7 years (Mean = 4.1 years). The participants’ proficiency was determined on the basis of the in-house proficiency placement test which they had already taken prior to enrolling at the institute where they studied. Furthermore, as for their educational level, the participants comprised junior high school (45%), high school (51%), and tertiary level (4%) students. Besides, none of the participants had left the country or lived abroad before.

3.2. Instruments

The data were gathered by means of a 54-item questionnaire. Some of the questionnaire items were adopted from Feyér (2012) and some others from Lefkowitz and Hedgcock (2006) while a few other items were developed by the researcher based on his experience and predictions of what might relate to accent and identity. The participants were required to respond to the items on a 7-point Likert scale (1: Strongly disagree, 2: Disagree, 3: Somewhat disagree, 4: Somewhat agree, 5: Agree, 6: Strongly agree, 7: Not reported or no experience). The contents of the questionnaire items were concerned with four major areas and, as a result, it was divided into four sections: (a) importance of pronunciation for communication (16 items), (b) factors influencing pronunciation (12 items), (c) pronunciation and (non)native speakers’ status (20 items), and (d) accent and its relation to identity (6 items).

3.3. Native vs. nonnative speakers

In the present study, a native speaker is defined as anyone who was born in an English-speaking country (where English is spoken as a first language) such as England, Canada, or the US, has learned English as his/her native language and has spoken English as his/her native language since childhood. On the other hand, a nonnative speaker is defined as an individual who was not born in an English-speaking country and has not learned English as his/her native language. Therefore, nonnative speakers include those who speak English as a second or foreign language. These terms were carefully explained to the participants before they embarked to fill out the questionnaires.

3.4. Procedure and data analysis

After the questionnaire had been constructed with the help of previous research, it was translated into Persian. The English-to-Persian translation was done by the researcher and was intended to ensure the participants’ full comprehension of the questionnaire items since it was assumed that the participants might have difficulty understanding the questionnaire statements in English. The translation was further examined by an expert in the field of translation for any inaccuracies or ambiguities and subsequent modifications were carried out. The researcher was present while the participants filled out the questionnaires and answered any ambiguities concerning the items. The participants were also reassured that their responses would be used solely for research purposes and that their information would be kept confidential. To analyze the data, the researcher reported means, modes, standard deviations, frequencies, and percentages to compare the participants’ responses.

4.1. Importance of pronunciation for communication

Table 2 presents the results of the participants’ attitudes toward the importance of pronunciation for communication which was addressed in the first research question.

It is seen from Table 2 that while the participants agree, whether slightly or strongly, with most of the questionnaire items concerned with the importance of pronunciation for communication, they seem to be more inclined toward some items such as items 1 (Strongly agree: 51%), 4 (Strongly agree: 74.5%), 5 (Strongly agree: 66.7%), 7 (Strongly agree: 70.6%), and 16 (Strongly agree: 64.7%). As can be seen, the highest level of agreement relates to item 4 which deals with the relationship between learners’ pronunciation and their confidence. On the other hand, the participants have expressed the highest level of disagreement with items 6 (Somewhat disagree: 11.8%), 12 (Somewhat disagree: 23.5%), and 14 (Strongly agree & Somewhat disagree: 19.6%). Finally, nearly the same level of disagreement and agreement, which can be interpreted as the participants’ uncertainty, can be seen in the case of item 12 which states that it is sufficient to understand the gist of the message with no regard to one’s accent or pronunciation.

4.2. Factors influencing pronunciation and accent

This part reports the results of the participants’ perceptions of the factors that impact on pronunciation and accent which was the subject of the second research question.

Table 3 shows that the most frequently selected response is “Somewhat agree” as can be seen from items 18, 20, 22, 24, and 25, followed by “Agree” in items 19, 23, 26, and 28. On the contrary, the highest level of disagreement is seen in item 17 which indicates 62.7% of overall disagreement, indicating that the participants do not perceive choral class repetition as an effective pedagogical technique. Nonetheless, the participants’ agreement with most other items might be indicative of the fact that all these factors (e.g., the presence of speakers from the same or opposite sex, peers with better oral skills, and repeating after the teacher/class) equally affect one’s pronunciation. This table also confirms that about one-third of the participants have no experience of the presence of the opposite sex in their classes (item 27), a familiar situation in language education in Iranian classes.

4.3. Pronunciation and native and nonnative speakers

Table 4 presents the participants’ perceptions toward native speakers’ and nonnative speakers’ pronunciation and their preference for each group. This issue was addressed in the third research question.

It is seen from Table 4 that the participants agreed most of all with items 29 (Strongly agree: 43.1%), 41 (Strongly agree: 33.3%), and 48 (Strongly agree: 62.7%). These items relate to the role of the instructor in providing the learners with native/nonnative pronunciation, sounding like NSs, and NSs as the best model of the English accent, respectively. On the other hand, the most disagreement is seen in the case of items 30 (Strongly disagree: 41.2%), 39 (Disagree: 25.5%), 40 (Strongly disagree: 27.5%), and 43 (Strongly disagree: 49%). Perhaps more interesting is the participants’ relatively high inexperience with regard to items 32 (47.1%) and 36 (29.4%) which are concerned with sounding best in the presence of NSs and feeling uncomfortable to sound like NSs in the presence of members from the opposite sex, respectively. Finally, the most dispersed responses belong to items 34, 44, 45, 46, and 47. It can be seen that in most of these items the level of agreement and disagreement is nearly equal, indicating the participants’ reservations about the issues raised in these items (i.e., anxiety as a result of the presence of one’s instructor or peers, dissatisfaction with Persian-accented speech or foreign accents other than Persian, and acceptability of having different pronunciations).

4.4. Accent and identity

The last research question enquired as to the participants’ perceptions of the relationship between one’s accent and identity. The results are displayed in Table 5.

Compared to Tables 2, 3, and 4, Table 5 shows the participants’ highest level of disagreement particularly in items 49 (Strongly disagree: 23.5%), 50 (Strongly disagree: 47.1%), 51 (Strongly disagree: 27.5%), and 54 (Strongly disagree: 23.5%). Besides, the participants’ agreement in this part is relatively low (see items 52 and 53). All in all, Table 4 indicates that the participants’ responses are quite dispersed, ranging from “Strongly disagree” to “Strongly agree.” All these indicate that the participants do not regard identity and socioeconomic and educational status as related. Two other items confirm this assumption, namely items 53 and 54. In this connection, item 53 demonstrates the participants’ remarkable agreement to the lack of relationship between one’s accent and his/her identity. In addition, the participants mostly refused to display their identities through their L1 accents (item 54).

4.5. Discussion

Current research stresses the significance of communication over any other aspect of speech including attractiveness of one’s accent. As Mey (1985, cited in Miller, 2004, 312) once argued, “one’s highest priority in speaking a foreign language should be to make oneself understood; sounding right is definitely a subordinate goal.” Despite Mey’s call for a communication-oriented goal in speaking the target language, the results of the present study indicate that the participants were strongly inclined to sound like native speakers rather than nonnative speakers. This finding is clearly in line with the results of previous similar research conducted in different contexts (e.g., Scales et al. 2006; Timmis 2002). An important finding is the participants’ confirmation that they enjoyed NSs’ accents, for example while watching movies, even when they did not understand everything being said. It is likely that EFL learners’ negative evaluations of NNSs’ accents are biased in favor of native speakers regardless of the fact that NNSs’ speech might also be as intelligible as, if not less intelligible than, that of NSs, an assumption that has already been supported by some researchers (e.g., Kim 2008). Likewise, EFL learners might hold such attitudes with the aim of avoiding the possible future negative evaluations, mockery, and racism on the part of NSs (see, e.g., Baugh 2000). Such biased, stereotypic attitudes may also exist against one group of NSs versus another group. For instance, Jarvella et al. (2001) have shown that Danish learners of English find British English more attractive than American English, both being NS. It can be argued that the prevalence of these attitudes results from the fact that pronunciation teaching has long been dominated by inner-circle norms and standards (Sifakis and Sougari 2005).

Another finding was the participants’ confirmation of the positive effect of “good” pronunciation on their confidence. Similarly, in the scholarly literature the ability to speak with a native accent has been associated with more self-confidence as well (Butler 2003). The participants sound the most dubious in their views of the factors impacting on pronunciation as “somewhat agree” is the most frequently selected response for six items (see Table 3). In addition, the participants’ substantiation of the ineffectiveness of repeating after the teacher as a technique for pronunciation instruction and their agreement with having better solitary pronunciation are of pedagogical significance (see Table 3). Although this is intended to be a tentative conclusion in want of experimental research, it is perhaps to the learners’ own advantage to allow them to practice and improve their oral skills such as pronunciation and accent on their own.

The results also indicated that nearly half of the participants reported that they did not have any experience of speaking in the presence of NSs while the others did not agree that their pronunciation sounds best in the presence of NSs. Such a standpoint might stem from the participants’ fear of NSs’ mockery of NNSs’ “funny accents” (Kubota 2001) in addition to fears of discrimination that is practiced against NNSs (Lindemann 2005; Lippi-Green 1994, 2012; Nguyen 1994). If this conclusion proves tenable, then it is plausible to discuss the participants’ views in light of Norton’s (1995, 1997) post-structuralist theory of social identity which attempts to elucidate the learner’s language learning experience in terms of power relations. According to Norton (1995), “SLA theory needs to develop a conception of the language learner as having a complex social identity that must be understood with reference to larger, and frequently inequitable social structures which are reproduced in day-to-day social interaction.” It is likely that the participants view NSs as having a top-down look at NNSs, a look in which NNSs are regarded as inferior to NSs due to having an accent when speaking English. This assumption sounds relatively plausible as accent is perceived to be a means of expressing one’s identity. In general, the results confirm the increased anxiety that learners feel when speaking the target language with a native accent as opposed to when putting on a foreign accent with their peers. It is fairly reasonable to hypothesize that viewing English as a commodity owned solely by its native speakers results in anxiety in learners because they deem themselves as incompetent, failed users of the target language, so to speak (see Aiello 2016). Furthermore, they reported that they felt anxious when speaking in the presence of speakers from the opposite sex unlike while speaking in the presence of speakers from the same sex which is totally understandable as nearly all language classes in private institutes do not follow a co-education policy.

What is more, the study presents useful insights into how the participants view accent as linked to identity (see Table 5). The participants’ dispersed responses in this part might be suggestive of their uncertainty and reservations as to whether and to what extent an individual’s identity is linked to and can be displayed by his/her socioeconomic and educational status. In general, the findings do not indicate the participants’ confirmation of any particular connection between accent and identity, a finding which supports Levis’s (2015) recent study. Arguably, in relation to identity, the participants’ marked preference for native-like pronunciation can hardly be interpreted as a possible means to negotiate their identities as legitimate speakers of English (Golombek and Jordan 2005). Rather, it can be viewed as their attempt to assimilate into the TL community and interact with NSs in more efficient ways. This finding is remarkably in line with Waniek-Klimczak, Rojczyk, and Porzuczek (2015) whose research findings indicated Polish learners’ unwillingness to display their ethnic identity through their Polish-accented pronunciation which the learners viewed as incorrect as a result of the presence of L1 features in their English accent. According to Turek (1990), such an orientation toward accent can demonstrate the learners’ integrative motivation and purposes of learning English. Researchers have, nonetheless, emphasized the need for the negotiation of learner identity. Drawing on Bucholtz and Hall (2005) work on identity, Macdonald (2015) argues that learners should engage in formulating their identities by positioning themselves as desirable speakers in interaction with other interlocutors. According to Macdonald (2015), this positioning helps learners develop confidence and therefore take active part in the language learning endeavor. As language is the individual’s most immediate and primary means to express one’s sociocultural identity (Sifakis and Sougari 2005), it is essential to examine how identity enactment is to be translated particularly in this age of rapid globalization.

5.1. Implications of the study

The participants’ fervent desire to imitate native speakers and to sound similar to them is very insightful, having clear pedagogical implications for practice. Primarily, it suggests that teachers, who, most probably, serve as the students’ most immediate input sources, need to improve their own pronunciation and must be equipped with the latest and most effective techniques of teaching and learning pronunciation. Second, still relevant to pronunciation instruction is the inherent implication for teacher education programs and policies and the attention they need to pay when it comes to developing teachers’ and students’ oral skills. Given that pronunciation instruction has long been a marginalized area of research and practice (Derwing and Munro 2005), most teachers are left to rely on their own intuitions and understanding of what constitutes efficient strategies to teach pronunciation. Perhaps most importantly, learners’ awareness needs to be raised as to the status of English in today’s world and as to the fact that not every deviation from Inner Circle norms is an error (Jenkins 1998; Low 2015); rather, it must be understood that all users of English, whether native or non-native speakers, have equal ownership over ELF (Aiello 2016).

5.2. Limitations of study

The findings of the present study are not conclusive given that the present study was limited from a number of aspects. First, it must be noted that in spite of the above arguments in favor of NNSs, some researchers perceive the NS model as better fitting educational purposes. Kuo (2006), for instance, states, “A native-speaker model […] would appear to be more appropriate and appealing in second language pedagogy than a description of English which is somewhat reduced and incomplete.” Besides, the participants included males only. Methodologically, quantitative research studies with a larger number of participants are required. Finally, qualitative data contribute substantially to understanding learners’ reasoning about accent and how it relates to identity in an EIL context.