2.1.1 Students in TESOL programs

It has become recurrent in the last few years to point out the ever-growing number of non-native speakers and learners of English in the world (e.g., Govardhan, Nayar & Sheorey Reference Govardhan, Nayar and Sheorey1999; Graddol Reference Graddol, Graddol and Meinhof1999, Reference Graddol2006; Crystal Reference Crystal2002, Reference Crystal2003). Braine (Reference Braine1999), J. Liu (Reference Liu and Braine1999a), Prodromou (Reference Prodromou2003), and Graddol (Reference Graddol2006) explain that as a result of these escalating demands in English instruction, the majority of trained ESL/EFL teachers in the world must be NNS teachers. Canagarajah (Reference Canagarajah2005) adds that 80% of the English teachers in the world are NNSs. These teachers are used to provide English instruction exclusively in EFL contexts, but now are found occupying teaching positions in English-speaking countries as well.

While many NNSs are trained and remain in their own countries, the number of those who go to English-speaking countries to attend TESOL MA programs is increasing. England & Roberts (Reference England and Roberts1989) surveyed foreign students and program administrators in TESOL MA programs in the United States. They found that about 40% of all TESOL students were NNSs of English, a number corroborated by D. Liu (Reference Liu and Braine1999).

In more recent years, Llurda's (Reference Llurda and Llurda2005b) study of 32 TESOL graduate programs in North America noted that 36% of the teacher trainees in those programs were NNSs of English. Of these NNSs, 78% had travelled to the US to attend the TESOL program and were likely to return to their countries after graduation. Polio's (Reference Polio1994) survey of 43 NNSs MA TESOL students also showed that 90% of them planned to return to their countries to teach English after obtaining their degrees.

In contrast, there are no similar studies conducted in the EFL context: Medgyes (Reference Medgyes and Braine1999) estimates that the majority of students in EFL TESOL programs are non-native speakers of English.

2.1.2 The content of TESOL training programs

One point of consideration is whether NNSs receive adequate teacher preparation, specifically addressing their needs. England & Roberts' (Reference England and Roberts1989) study showed that none of the 123 Master's programs they surveyed offered additional or different training for NNSs, mostly because of lack of research in this field as well as limited financial resources of the schools or departments. Curiously, program administrators recognized linguistic and cultural differences between NSs and NNSs yet did not see a need for special adjustments to accommodate NNSs' needs.

Flynn & Gulikers (Reference Flynn and Gulikers2001) explain that in order to better prepare teachers to teach in Intensive English Programs (IEPs), MA TESOL programs should offer courses in both applied linguistics and curriculum design, require all student teachers to do a practicum, and give them the opportunity to observe and teach in different contexts (primary and secondary schools, community colleges, IEPs, etc.). However, a survey of 194 TESOL programs in North America by Govardhan et al. (Reference Govardhan, Nayar and Sheorey1999) shows that several of the courses taught in these programs relate only vaguely to English teaching and in particular to teaching English in EFL contexts. Similarly, Polio & Wilson-Duffy (Reference Polio and Wilson-Duffy1998) note that few courses in MA TESOL programs address the issues and challenges facing NNSs student teachers planning to go back to their countries to teach English. In the same vein, several scholars (e.g., Braine Reference Braine1999; D. Liu Reference Liu and Braine1999; Oka Reference Oka2004; Canagarajah Reference Canagarajah2005; Holliday Reference Holliday2005) suggest that the theoretical and pedagogical courses taught in North American TESOL programs do not always correspond to what is needed in EFL contexts.

Kamhi-Stein (Reference Kamhi-Stein and Braine1999) argues for the inclusion of specific issues related to non-native speakers in language teacher education programs as a way of empowering non-native English speakers of TESOL. Cullen (Reference Cullen1994) emphasizes the fact that both NSs and NNSs could benefit from being taught a course in grammar, pronunciation, vocabulary, and culture. Such courses would help future teachers in the areas of vocabulary building, pronunciation, culture, and general fluency (Lee Reference Lee and Kamhi-Stein2004; Lin, Wang, Akamatsu & Riazi Reference Lin, Wang, Akamatsu, Riazi and Canagarajah2005). They would also teach the value of collaboration of native and non-native English-speaking teachers, show how to take advantage of their respective strengths and weaknesses (de Oliveira & Richardson Reference de Oliveira and Richardson2001, Reference de Oliveira, Richardson and Kamhi-Stein2004; Matsuda & Matsuda Reference Matsuda, Matsuda and Kamhi-Stein2004; Gebhard & Nagamine Reference Gebhard and Nagamine2005), and sensitize native speakers of English to issues such as culture shock, language learning difficulties, and other intercultural and sociolinguistic issues (Perdreau Reference Perdreau1994; Reid Reference Reid1997; Brady & Gulikers Reference Brady, Gulikers and Kamhi-Stein2004; Pasternak & Bailey Reference Pasternak, Bailey and Kamhi-Stein2004).

Llurda (Reference Llurda and Llurda2005b) shows that although practicum supervisors agreed that their NNS student teachers had higher language awareness than NSs, most of them also said that they would recommend NNS to teach primarily low-level classes. Similarly, most practicum supervisors thought their NNSs would feel more comfortable teaching in their own countries rather than in the US. In contrast, many practicum supervisors acknowledged that their program was highly competitive and only accepted the ‘cream of the crop’, that some of their NNS students were very bright, and that NNSs brought diversity to the program.

Recognizing the assets NNSs bring to TESOL programs, Kamhi-Stein (Reference Kamhi-Stein2004), corroborating Brinton's (Reference Brinton and Kamhi-Stein2004) findings and suggestions, emphasizes the idea that TESOL programs tailored to the needs of the NNSs of English would increase the future teachers' motivation and therefore their self-esteem. Such programs would allow NNSs to develop an understanding of their own assets, values, and beliefs, and ‘promote an improvement of the teacher-trainees competencies’ (Kamhi-Stein Reference Kamhi-Stein2004: 148). As Brinton explains, ‘[t]eacher educators have the responsibility to assist [NNS student teachers] in gaining a sense of self as teacher, in creating an atmosphere that fosters respect, and in providing guidance in culturally appropriate norms of behavior and discipline’ (Brinton Reference Brinton and Kamhi-Stein2004: 202).

Moussu (Reference Moussu2006) asked 96 native and non-native English-speaking ESL teachers about their overall teacher preparation. Responses showed that several teacher education programs did not allow NNSs to do student teaching (teaching practicum) either in an Intensive English Program attached to the TESOL MA program or at a separate school. This corroborates Reid's (Reference Reid1997) and Mahboob's (Reference Mahboob2003) observations about the lack of practical training available to many NNSs. Brady & Gulikers (Reference Brady, Gulikers and Kamhi-Stein2004) clearly explain that even when positions are available for student teachers to do their practicum, many teacher educators hesitate (or in many cases refuse) to let NNSs apply, fearing that NNS teachers' lack of experience or poor linguistic skills impede the ESL students' learning.

In EFL settings, the situation seems to be quite similar. Medgyes (Reference Medgyes and Braine1999) reports on the Center for English Teacher Training in Budapest, Hungary, where most administrators do not see the need for additional instruction for NNSs of English. NNSs, however, demand additional classes in pronunciation and vocabulary. Berry (Reference Berry1990), Nelson (Reference Nelson1992), Eguiguren (Reference Eguiguren2000) and Llurda (Reference Llurda2004) support this argument and hope for classes that would increase NNSs' confidence in their language and teaching skills, and as a consequence would facilitate language use in the classroom, address the learners' needs and diverse backgrounds, and present different methodologies to student teachers who intend to teach in other countries of the world and may not necessarily want to learn a monocentric teaching methodology.

Finally, Giauque (Reference Giauque1984) felt that NNSs were not the only ones who could become better teachers with better preparation. He explains that even though it is imperative for NNSs to acquire a good knowledge of the language, it is equally essential that NSs gain a good knowledge of contrastive linguistics before being qualified to teach their own language. This argument is supported by Rampton (Reference Rampton1990), who states that being born into a language does not mean that one inherently speaks it well. Seidlhofer (Reference Seidlhofer1999) complements this view by using the following metaphorical image: ‘native speakers know the destination, but not the terrain that has to be crossed to get there; they themselves have not travelled the same route’ (238).

2.2.1 As seen by others

Some of the first reflections regarding the differences between native and non-native speaking ESL/EFL teachers came in the eighties (e.g., Kachru Reference Kachru and Kachru1981; Pride Reference Pride1981; Nickel Reference Nickel1985; Coppieters Reference Coppieters1987; Kresovich Reference Kresovich1988). Edge (Reference Edge1988), for example, advocated for the importance of giving ‘real’ models (that is, native speakers of the ESL/EFL students' languages) to the students. These ‘real’ models speak the language of the students natively and have learned to speak English well, as opposed to the ‘foreign’ models (NSs), who do not share the cultural, social, and emotional experience of the students, a perspective shared later on by McKay (Reference McKay2003).

In the early nineties, Medgyes wrote the first article (Reference Medgyes1992) comparing native and non-native English-speaking teachers, in which it was stated that

• • the ideal NS teacher is the one who has achieved a high degree of proficiency in the learners' mother tongue;

• • the ideal NNS teacher is the one who ‘has achieved near-native proficiency’ in English (Medgyes Reference Medgyes1992: 348f.).

Canagarajah (Reference Canagarajah and Braine1999) added that NSs will be better teachers in EFL contexts, because of their unique cultural knowledge, whereas NNSs will be better teachers in ESL contexts, because of their multicultural experience. Interestingly, as seen earlier, this claim is not supported at all by TESOL practicum supervisors, who seem to believe that NNS teachers would be better teachers in their own countries (Llurda Reference Llurda and Llurda2005b).

In a later discussion about NNS teachers' advantages and disadvantages, Medgyes (Reference Medgyes1994) described six positive characteristics: 1) They provide a good learner model to their students; 2) They can teach language strategies very effectively; 3) They are able to provide more information about the language to their students; 4) They understand the difficulties and needs of the students; 5) They are able to anticipate and predict language difficulties; and 6) In EFL settings, they can use the students' native language to their advantage. Medgyes then explains that if the language ‘deficiencies’ of the NNS teachers are remedied, native and non-native English-speaking teachers have equal chance to achieve professional success.

After these reflections and discussions, several studies were conducted that investigated the linguistic and pedagogical differences that may exist between native and non-native English-speaking teachers. In comparing expert and novice NSs and NNSs in an EFL context, for example, McNeill (Reference McNeill and Llurda2005) noticed that novice Chinese NNS teachers were very skilled at predicting which words would be easy and difficult to understand for Cantonese-speaking EFL students. Conversely, both expert and novice NS teachers were quite incapable of making accurate predictions. Following Ellis' (Reference Ellis2004, Reference Ellis2006) arguments, this might have to do with monolingualism vs. bilingualism rather than native or non-native status of such teachers.Footnote ² When it comes to finding and correcting errors, although Sheorey (Reference Sheorey1986) claimed that NNSs were often less tolerant of errors than NSs when grading college-level ESL compositions, Porte (Reference Porte1999a) discovered that NSs and NNSs error judgments in an EFL context were rather similar, and he suggested that NNSs living and working overseas for an extended period might incorporate some of the features typically found among NNS teachers (Porte Reference Porte1999b).

Regarding language awareness, Barratt & Kontra (Reference Barratt and Kontra2000) showed that NS teachers could also easily discourage their students since they are rarely able to make useful comparisons and contrasts with the learners' first language. Additionally, according to Barratt & Kontra, NSs are often unable to empathize with students going through the learning process.

Árva & Medgyes' (Reference Árva and Medgyes2000) results corroborate those obtained by Barratt & Kontra (Reference Barratt and Kontra2000). Their study shows that a unique advantage NNS English teachers have over NS teachers is that they can empathize very well with their students' learning difficulties and understand what it is to be homesick and to experience culture shock (in ESL contexts). Finally, and very importantly, NNS teachers can be greatly admired by their students because they are successful role models and are often very motivated (Lee Reference Lee2000). As Cook (Reference Cook and Llurda2005: 57) explains, NNS teachers ‘provide models of proficient [L2] users in action in the classroom [and also] examples of people who have become successful [L2] users’. That is, NNSs demonstrate to their students what it is possible to do with a second language and their appreciation for that language and its culture.

In a study that asked US school host teachers for their opinions about the NNS student-teachers who were assigned to their class for their practice teaching, Nemtchinova (Reference Nemtchinova2005) described the NNS trainees as having usually a good contact with their ESL students, using correct, authentic, and fluent English in general, and being very well prepared for class although they did not always make good use of class time. The NNS trainees in her study also seemed to build good relationships with their students, understood their needs, communicated well with them, and gave them positive feedback and fair evaluations. Finally, NNS trainees usually seemed to have a sufficient cultural awareness of the target culture and language context. Several host teachers noted, however, that many of the trainees self-evaluated themselves very harshly and sometimes lacked self-confidence.

Looking at the issue from a different perspective, P. K. Matsuda (Reference Matsuda1997) and Matsuda & Matsuda (Reference Matsuda and Matsuda2001) explain that rather than comparing competencies or deficiencies, an ideal model for both native and non-native English-speaking teachers would look at all teachers as a ‘cooperative learning community and consider their development holistically’ (P. K. Matsuda Reference Matsuda1997: 13). Instead of looking at NSs and NNSs as two distinct groups, one necessarily better or more qualified to be teachers than the other, Matsuda & Matsuda emphasize cooperation and mutual help between NS and NNS teachers, since both groups of teachers have specific advantages and weaknesses. However, in spite of the existence of successful research experiences on NSs and NNSs cooperation, such as de Oliveira & Richardson (Reference de Oliveira and Richardson2001, Reference de Oliveira, Richardson and Kamhi-Stein2004), they are still very difficult to find in the literature.

2.2.2 As seen by themselves

Several scholars have asked non-native teachers, student-teachers, and teacher educators directly for their opinions and self-perceptions of their strengths and weaknesses. For example, Reves & Medgyes (Reference Reves and Medgyes1994) conducted a study that showed that the perpetual fear of their students' judgment made the participating 216 non-native English-speaking EFL teachers feel constantly self-conscious of their mistakes. According to Reves & Medgyes' participants, this ‘self-discrimination’ often leads to a poorer self-image, which further deteriorates language performance, which, in turn could lead to an even stronger feeling of inferiority.

This point of view may seem extreme, and yet other language teachers, new teachers of all languages, or any teacher with poor self-esteem, might experience similar feelings. It is interesting to notice, however, that it seems acceptable for NS teachers to make some occasional mistakes while teaching, or not to know all the details about the English language (Amin Reference Amin and Kamhi-Stein2004). In contrast, when NNS teachers make the same mistakes or do not know everything about the English language, their teaching abilities and competencies are often immediately questioned (Canagarajah Reference Canagarajah and Braine1999, Reference Canagarajah2005). This attitude from the students, NS colleagues, and often even from the NNS teachers themselves, will often lead to feelings of inadequacy and self-doubts (Braine Reference Braine and Kamhi-Stein2004; Morita Reference Morita2004).

In another study about self-perception, Samimy & Brutt-Griffler (Reference Samimy, Brutt-Griffler and Braine1999) investigated how seventeen non-native English-speaking TESOL graduate students perceived themselves as future NNS teachers. While all the participants were currently students in a TESOL program, several of them were, or had been, teaching ESL or EFL for a number of years. The respondents seemed to be aware that factors such as the age and level of the students, the goals and objectives of the program, and the personality and teaching skills of the teachers made a significant difference in how successful a teaching/learning experience could become.

Participants to the Samimy & Brutt-Griffler (Reference Samimy, Brutt-Griffler and Braine1999) study also felt that it was sometimes harder for them to feel qualified and appreciated in an ESL context, where their competences are more often questioned. In contrast, they thought it easier to see themselves as role models ‘in social, cultural, emotional, or experiential terms’ (138) and to be valued and respected as professionals when teaching in their own countries.

Amin (Reference Amin1997) interviewed five ‘visible minority’ women about their teaching experiences in Canada. These women believed their students thought that only Caucasian teachers could be native speakers of English. They also believed that only Caucasian native speakers of North American English could know ‘real’ and ‘proper’ English. Consequently, those teachers felt constantly judged and compared with native, white, teachers (see Holliday Reference Holliday2008 for similar concerns). Gender also seemed to be a serious issue for women teachers who have difficulties establishing their authority. According to Amin, ESL students' referent thus seems to be a white, native-English-speaking Anglo male. This attitude towards ‘whiteness’, as well as its resulting conflicts with identities and legitimacy, was also felt by Golombek & Jordan's (Reference Golombek and Jordan2005) interviewees.

A few other studies were conducted on teachers' self-perceptions. J. Liu (Reference Liu and Braine1999a, Reference Liu and Llurda2005) investigated the thoughts and perceptions of university ESL faculty members and graduate students responsible for the teaching of undergraduate courses. His results show that their teaching experiences were affected by: the level of students (graduate students respected and admired NNSs more than undergraduates did), the race and accent of the teachers, the course the NNSs were teaching, and the teachers' individual teaching methods. Similar results were obtained by Tang (Reference Tang1997) and Moussu (Reference Moussu2002, Reference Moussu2006) (see also Moussu & Braine Reference Moussu and Braine2006), who found that non-native ESL and EFL teachers experienced different reactions from students coming from different countries.

Maum (Reference Maum2003) asked 80 primary and secondary school teachers about their beliefs and experiences as native and non-native ESL teachers in adult education. Her study showed that NNS teachers, more than NS teachers, found the ESL teachers' cultural background and training in linguistics to be very important, and also that including cross-cultural issues into the teaching of ESL had much value. Surprisingly, the NS teachers in this study were not aware of any discrimination against NNS teachers, while NNS teachers clearly expressed their frustration towards their isolation and ‘marginalization in the profession’ (162).

In another study, Kamhi-Stein, Aagard, Ching, Paik & Sasser (Reference Kamhi-Stein, Aagard, Ching, Paik, Sasser and Kamhi-Stein2004) asked 55 native English-speaking and 32 non-native English-speaking primary and secondary school teachers about their self-confidence in speaking and teaching English. Both NSs and NNSs seemed to be confident in their language skills, although NNS teachers' responses were slightly less positive than NS teachers'. A surprising result was that NNSs did not rate their pronunciation and communication skills as negatively as expected, while grammar was not ranked as NNSs' strongest skill, a result contradicting previous findings but later corroborated by Moussu (Reference Moussu2006), who followed up on Kamhi-Stein et al.'s (Reference Kamhi-Stein, Aagard, Ching, Paik, Sasser and Kamhi-Stein2004) study.

In a very different context (Spain), Llurda & Huguet (Reference Llurda and Huguet2003) asked 101 non-native English-speaking EFL teachers in primary and secondary schools about their perceived language skills, pedagogical skills, and views on issues related to the NS–NNS teacher debate. Secondary teachers perceived their English skills as overall higher than those of primary teachers, and they held a more critical position regarding the NS–NNS teacher debate, expressed through a more supportive attitude towards NNS teachers and a lower dependence on the native speaker as the ultimate model in language teaching. In the same context, Llurda (forthcoming) found that length of time spent in English-speaking countries was a significant factor in determining NNS teachers' self-perceptions. Supporting Dalton-Puffer, Kaltenboeck & Smit's (Reference Dalton-Puffer, Kaltenboeck and Smit1997) results pointing to a more critical attitude on teachers with long stays abroad, Llurda (forthcoming) found that NNSs teachers who had never or hardly ever stayed in English-speaking countries were more eager to support the native speaker as the ideal teacher, and British English as the variety they would like their students to end up speaking, as opposed to American English, or International English.

In another EFL context (Greece), Sifakis & Sougari (Reference Sifakis and Sougari2005) asked 421 teachers about pronunciation issues such as their attitude towards their own accent or the frequency of feedback on pronunciation. They found differences in the frequency of pronunciation feedback given by NNS primary and secondary teachers, and they additionally observed that the adoption of a model of English pronunciation according to the notion of English as an International language (Jenkins Reference Jenkins2000, Reference Jenkins2007) was not considered.

In Turkey, Dogancay-Aktuna (forthcoming) asked 21 non-native English-speaking teacher educators about their status as non-native speakers of English, professional identities, and self-perceived skills. Most of these participants rated their language skills and competences in English as high, overall, although some noted a need to improve their knowledge of idiomatic expressions and conversational English. At the same time, slightly more than half of the participants had experienced prejudice because of their non-native status and many felt that this status was disadvantageous to their professional careers and teaching experience. They agreed, however, that being NNSs in an EFL context allowed them to understand the issues related to this context better than if they were NSs of English. Bayyurt (Reference Bayyurt2006), who interviewed 12 Turkish NNS teachers about their beliefs regarding the teaching of culture in the EFL classroom, additionally showed that NNS teachers were aware that EFL students regarded them as good language learning models and guides.

In the Japanese context, Butler (Reference Butler2007) investigated Japanese elementary school teachers' attitudes towards the privileged status of NS English teachers and their self-evaluations of their English proficiency. First, she found out that approximately 60% of her 112 respondents supported the notion that native speakers of English were the best ESL/EFL teachers and only 13% did not. These teachers also believed that ‘standard English’ only (British and American English) should be taught to EFL students. Second, her respondents self-evaluated themselves as having stronger reading skills than writing and oral (fluency, grammar, and vocabulary) skills. Interestingly, the teachers who believed they had the lowest English proficiency were also those who most strongly believed that English was best taught by NSs.

A different approach to NNS teachers' self-perceptions was taken by Ellis (Reference Ellis2002, Reference Ellis2004), who focused on NNS teachers in Australia as multilingual speakers and the advantages reported by their multilingual condition. Her findings show that both NSs and NNSs multilingual teachers have ‘more in common with each other than with the monolingual teachers’ (Ellis Reference Ellis2004: 96), an element that, as the author points out, has several implications for further research into the NS–NNS dichotomy.

An overall pattern is apparent in all the studies discussed above, as it becomes clear that if NS and NNS teachers perceive their strengths and weaknesses differently, it is not always clear as to how and why.

4.1.1 Non-empirical reflections on the nature and conditions of non-native speakers

Some of the key concepts that would later characterize the literature on NNS teachers were already present in the earlier writings by Medgyes (Reference Medgyes1983, Reference Medgyes1986), and constituted a central element in his book (Medgyes Reference Medgyes1994), which set the basis for more extensive research, and gave the topic a new status as a subject suitable for scientific investigation.

Looking back at the research methods used, it is quite striking to observe that a large number of studies are based on non-empirical think-pieces that stress the need to reconsider the role of non-native language teachers and to value their important contribution to the teaching of languages worldwide (e.g., Greis Reference Greis1984; Seidlhofer Reference Seidlhofer1996; Amin Reference Amin1997; Braine Reference Braine1999; Brutt-Griffler & Samimy Reference Brutt-Griffler and Samimy1999; Lee Reference Lee2000; Llurda Reference Llurda2004; Derwing & Munro Reference Derwing, Munro and Llurda2005; Modiano Reference Modiano and Llurda2005; Rajagopalan Reference Rajagopalan and Llurda2005).

As important as these pieces of writing have been for the development of a research agenda on issues related to NNS teachers, the excessive reliance on this kind of work poses a clear danger to the field, with the likely scenario of authors running out of new views and the field ending up with an inflationist repetition of the same ideas in different words and by different authors. As necessary as it still may be to raise awareness on the importance of being a non-native teacher (Seidlhofer Reference Seidlhofer1996), the field should not stay forever wondering about the need to increase NNSs' prestige and professional self-esteem, if we want to consolidate the area as one of the most active current research topics in the language teaching profession (Bailey Reference Bailey2001). It is only natural, then, that research on non-native teachers is becoming more sophisticated and ambitious in trying to further explore the characteristics of this group of teachers, as well as exploring new lines and new methods of research. Failing to do so would only lead to an abandonment of interest in NNS teachers by the applied and educational linguistics research community.

4.1.2 Personal experiences and narratives

Besides theoretical thinking and analysis of characteristics of non-native teachers, some researchers have centred their work on sharing their personal experiences as successful non-native teachers. These narrative accounts (e.g., Braine Reference Braine1999, Reference Braine2005; Connor Reference Connor and Braine1999; Thomas Reference Thomas and Braine1999; de Oliveira & Richardson Reference de Oliveira and Richardson2001, Reference de Oliveira, Richardson and Kamhi-Stein2004) have typically been intended to become sources of inspiration for other non-natives, who can see their own personal experiences reflected more clearly in such an account than in a set of figures coming from a more formally complex empirical study. Narratives are an excellent way to describe a situation or experience, giving it full internal coherence, in such a way that the narrated story becomes a transforming reality in itself as the epitome of all other untold stories shared by all readers who can see themselves reflected in it (Pavlenko Reference Pavlenko2007). The role of narratives as a way of establishing a tradition, and ultimately creating reality was clearly discussed by Bruner (Reference Bruner1991), who compared their explanatory power to that of positivist scientific research methods:

This approach to NNS teachers' reality, or rather, this way of constructing the reality of NNS teachers, has given us a few individual stories which have become the ‘tokens of broader types’ (Bruner Reference Bruner1991: 6). Although some of these narratives have appeared in non-research publications, others have made their way into more research-oriented environments (e.g., Braine Reference Braine2005; Bayyurt Reference Bayyurt2006). The adoption of narratives as a research mechanism constitutes a very positive outcome in language teaching research. However, research on NNS teachers still needs widespread recognition, and to become well established by the research community it needs to be complemented by more studies using quantitative data collection methods which follow the empirical conventions established in the social sciences.

4.1.3 Surveys

Once researchers interested in the development of current understanding of NNS teachers moved beyond the theoretical personal analysis of their professional situation, it became obvious that more empirical data was needed. The subjective nature of previous research needed the complement of more objective accounts, based on quantitative approaches that provided numerical data that could help legitimize the scientific nature of this area of study.

Surveys have proved to be a very popular method of research since Reves & Medgyes (Reference Reves and Medgyes1994) asked NNS teachers about their self-perceptions. Many other studies have later resorted to the use of surveys (e.g., Cheung Reference Cheung2002; Moussou Reference Moussu2002; Llurda & Huguet Reference Llurda and Huguet2003; Mahboob Reference Mahboob2003, Reference Mahboob and Kamhi-Stein2004; Kamhi-Stein et al. Reference Kamhi-Stein, Aagard, Ching, Paik, Sasser and Kamhi-Stein2004; Llurda Reference Llurda and Llurda2005b; Nemtchinova Reference Nemtchinova2005; Bayyurt Reference Bayyurt2006; Moussu Reference Moussu2006; Dogancay-Aktuna forthcoming), certainly up to a point where it may be agreed that a certain saturation of questionnaires and surveys has already been reached. One of the agreed problems with surveys is the extraordinary facility with which one may come up with a questionnaire and implement it with a ready-made group of participants. As Dörnyei (Reference Dörnyei2003) explains, it is rather easy to prepare a questionnaire, but preparing and implementing a good questionnaire is a much more difficult task. Dörnyei provides some basic elements that are needed for the design of questionnaires that can contribute to research in the social sciences by moving beyond the simplicity of asking a set of ad-hoc questions to a group of people available to the researcher. Still, it must be acknowledged that one of the main values of research on non-native teachers based on this method has been its capacity to document aspects that previously had only been pointed out through reflection and reports on personal experiences. Questionnaires also allow to report on very large numbers of participants and to duplicate studies easily, and therefore, they must be credited for providing the first empirical accounts on the nature and perceptions regarding non-native language teachers: as seen by students (e.g., Cheung Reference Cheung2002; Lasagabaster & Sierra Reference Lasagabaster and Sierra2002; Moussu Reference Moussu2002; Higgins Reference Higgins2003; Benke & Medgyes Reference Benke, Medgyes and Llurda2005; Lasagabaster & Sierra Reference Lasagabaster, Sierra and Llurda2005; Pacek Reference Pacek and Llurda2005; Moussu Reference Moussu2006) by program administrators and supervisors (e.g., Mahboob Reference Mahboob2003; Llurda Reference Llurda and Llurda2005b; Nemtchinova Reference Nemtchinova2005; Moussu Reference Moussu2006); or by NNS teachers themselves (e.g., Reves & Medgyes Reference Reves and Medgyes1994; Samimy & Brutt-Griffler Reference Samimy, Brutt-Griffler and Braine1999; Inbar-Lourie Reference Inbar-Lourie2001; Cheung Reference Cheung2002; Liang Reference Liang2002; Llurda & Huguet Reference Llurda and Huguet2003; Mahboob Reference Mahboob2003; Maum Reference Maum2003; McKay Reference McKay2003; Kamhi-Stein et al. Reference Kamhi-Stein, Aagard, Ching, Paik, Sasser and Kamhi-Stein2004; Inbar-Lourie Reference Inbar-Lourie and Llurda2005; Moussu Reference Moussu2006; Llurda forthcoming).

A more careful analysis of those questionnaires, however, would show that many of them do not comply with some of the requirements determined by Dörnyei (Reference Dörnyei2003). Reves & Medgyes (Reference Reves and Medgyes1994), for instance, distributed their questionnaires among representatives of The British Council in different countries on the five continents who were in turn asked to distribute them among both native and non-native teachers. The fundamental breakthrough of that study in opening the field of applied linguistics to research on non-native professionals notwithstanding, it is now obvious that such a method incorporated some flaws, particularly when one looks at such issues as the selection of participants. Current surveys tend to control the identity of participants in an attempt to obtain more reliable and valid results (e.g., Lasagabaster & Sierra Reference Lasagabaster and Sierra2002; Maum Reference Maum2003; Benke & Medgyes Reference Benke, Medgyes and Llurda2005). This effort to reach higher degrees of objectivity is not incompatible with the tendency in some recent questionnaires to incorporate open-ended questions (e.g., Mahboob Reference Mahboob2003; Maum Reference Maum2003; Lasagabaster & Sierra Reference Lasagabaster, Sierra and Llurda2005; Moussu Reference Moussu2006), which allow respondents to express their views on the matter without the constraints posed by closed questions previously established by the researcher, with no margin for respondents to incorporate their own intuitions and perceptions.

4.1.4 Interviews

Another fruitful method that has been implemented in research on non-native language teachers is based on the use of personal interviews, which have proved very rich in providing insights into the minds of NNS teachers and related people. The following extract from Holliday (Reference Holliday2005) is an excellent example of the great explanatory power of interviews, which convey a direct view of the minds of the subjects under study:

Interviews with NNESTs have been conducted in different formats: face-to-face, by email, or by telephone. In all cases, they have yielded data that is more complex and deeper than simple responses to questionnaires, even when these may have included open-ended questions. The use of interviews, either face-to-face or via email, allows researchers to construct narratives based on the reports of participants, who thus lend their own words to the shared construction of non-native language teachers' identity (e.g., J. Liu Reference Liu and Braine1999a, Reference Liub; Tsui Reference Tsui2003; Ellis Reference Ellis2004; D. Liu Reference Liu2004; Reference Liu and Llurda2005; Morita Reference Morita2004), but others have also used it as a complement to questionnaires (e.g., Kachi & Lee Reference Kachi and Lee2001; Cheung Reference Cheung2002; Holliday Reference Holliday2005; Bayyurt Reference Bayyurt2006).

Although interviews are somewhat qualitative in nature, this must be no obstacle for a certain degree of objectivity being pursued by studies using such methods. Thus, researchers should be very serious about not exerting any influence on subjects' responses, and offering guarantees that the reported responses freely and spontaneously originated from the participants, with no conditioning restrictions by the researcher. Unfortunately, sometimes this is not very easy to achieve. For example, Samimy & Brutt-Griffler's (Reference Samimy, Brutt-Griffler and Braine1999) study had a problem with participant selection: The participants were 17 TESOL NNS graduate students attending an optional seminar with the title ‘Issues and concerns related to NNS professionals’, offered within their program by one of the researchers, with a list of readings that may have likely affected students' views. Furthermore, the interviews took place ‘throughout the academic quarter depending on the availability of the participants’ (Samimy & Brutt-Griffler Reference Brutt-Griffler and Samimy1999: 132), with no further indication as to when each of the interviews were conducted, whether at the beginning or at the end of the semester, a variable that may certainly have had some effect on the responses given by the participants, as the readings and the discussions that were taking place during the course would gradually affect the students' views and their responses.

Another study that made use of interviews was Llurda (Reference Llurda2003), and the issue of participant self-selection was also problematic: the interviews were conducted among practicum supervisors who had previously responded to a written questionnaire in which they were asked to compare native speakers and non-native speakers on their MA programs. One may wonder what the responses to the interviews would have been if they had been conducted among supervisors who did not respond to the questionnaire. It is reasonable to expect more positive attitudes towards non-native teachers among those who devoted some of their time to responding to the questionnaire than those who did not.

These two examples illustrate the need for more interview studies in which the population is strictly controlled, so as to avoid some of the intervening factors that may have affected this type of studies in the past. Only then will personal narratives be seen to transmit a reality that lies at a deeper level than what can be accounted by the sheer use of figures and tables provided by survey-based quantitative studies.

4.1.5 Classroom observations

Very little research has actually looked at what happens within the language classrooms involving NNS teachers. We will refer here to three studies that focus on classroom performance with the aim to learn more about non-native teachers and their unique contributions to the language teaching profession. These studies are, however, very different from each other, as the focus of the first one (Árva & Medgyes Reference Árva and Medgyes2000) was to compare five native and five non-native teachers teaching in Hungarian secondary schools, first by using interviews that showed teachers' self-perceptions, and then by analysing one class taught by each of the ten participants in the study. The results were dealt with in a rather broad manner, mostly by commenting on the differential behaviours between natives and non-natives, and concluding that each of the two groups may be better qualified to teach different types of classes. Cots & Díaz (Reference Cots, Díaz and Llurda2005), on the other hand, conducted a micro-analysis on how four NNS teachers in Catalonia constructed social relationships and linguistic knowledge in the classroom through the use of teacher talk. Finally, Morita's (2007) study also used classroom observation (in addition to interviews and self-reports) to investigate NNSs' perceptions of linguistic identity and participation in their predominantly English-speaking classes.

The shortage of studies using a classroom-observation method points to an urgent need for more studies into NNS teachers' classroom performance, as we need to know more about their use of teacher talk, grammar explanations, promotion of varied interactional patterns, use of the textbook, and all the many specific NNS ‘characteristics’ that have been mentioned in the literature (cf. Medgyes Reference Medgyes1994; Reves & Medgyes Reference Reves and Medgyes1994; Tang Reference Tang1997; Seidlhofer Reference Seidlhofer1999).

4.2.1 Diversity within non-native English-speaking teachers

Research so far has typically approached NNSs as a single group, as though they could all be described through a single account. This may be partially due to the nature of the term we have been using in describing NNSs (i.e., non-native), which does not identify any particular characteristic of this group except for the negation of their native speaker condition.

Two problems can be observed here. One is the use of a negative particle to claim an identity, or better a ‘non-identity’. The second problem is the fact that the identity they claim not to have (i.e., native) is a particularly elusive one, as Paikeday (Reference Paikeday1985), Rampton (Reference Rampton1990), J. Liu (Reference Liu and Braine1999a), Brutt-Griffler & Samimy (Reference Brutt-Griffler and Samimy1999), Davies (Reference Davies1991, Reference Davies2003), and Faez (Reference Faez2007) have extensively discussed. No feature, other than birth within a fairly homogenous linguistic community, has been discovered to support the existence of a ‘native speaker’ identity. Ultimately, what appears to be the most distinguishing feature is simply whether one considers herself a native speaker of a given community and is recognized as such by other speakers in the community. A concept that is so elusive to characterize, and which has been so loosely applied to all speakers who would not meet the rather subjective and discriminatory criteria to belong to the ‘native speaker’ category, is prone to suffer from overgeneralization. As a consequence, there have been several studies which have dealt with the broad category of NNSs without any further detailing of the particular conditions and settings of that ‘group of speakers’, and without any distinction of different circumstances and characteristics within the NNS constituency.

Holliday (Reference Holliday2005, Reference Holliday2008) gives a detailed account of how NNSs are categorized as the ‘Other’ by the dominant group of NSs, and how this ‘otherness’ entails a necessary simplification and suppression of complexities similar to what happens in other instances of what Holliday names as ‘culturism’, the most common examples of which are found in the generalization of the other, typical of racism and sexism. Culturism, and more particularly ‘native speakerism’, creates a stereotypical myth by which the ‘other’ (i.e., the NNS) is seen as ‘uncritical, static, rigid, with a fixed view of knowledge, intellectually interdependent, wishing to preserve knowledge, good at memorizing’, who also needs ‘to be trained, treated sensitively, understood, involved, given ownership, empowered’, finds ‘decision-making difficult’, and prefers ‘frontal teaching’ as she is ‘exam-oriented’ (Holliday Reference Holliday2005: 21).

Differences do obviously exist within the NNS group, but they have been disregarded, probably for different reasons, by both NSs and NNSs. NSs have most likely fallen within the trap of ‘native speakerism’ by perceiving just one single entity of NNSs, instead of many. NNSs have probably disregarded differences in their attempt to claim their status and their role in the profession, a goal more easily achievable from a unified perspective than through the emphasis on internal differences. Obviously, native speakers do have differences among themselves, but this has been taken for granted by the research community as it is commonly assumed that different (NS) teachers have different ways of teaching. It is true that the literature on NNS teachers has also tried to characterize NS teachers as a group, probably as a reaction or need to establish a parallel contrast between the two. Our perspective, however, is that we cannot generalize and there is a need to deepen our knowledge of language teaching and how different factors among individual teachers may affect their performance.

We are perfectly aware it would be a temerity for us to undergo here a classification of the subgroups to be found within the NNS constituency, and we see the risk of establishing our own new generalizations if we pursued that goal. However, we think we may clarify what we mean when we claim the existence of differences within the NNS group by mentioning just a few that can be easily accepted, and which can benefit from further research.

One such difference is created by the setting where teachers develop their professional activity: whether the language taught is present in the social environment or not. In the tradition of English language teaching, these two settings are called ESL (when the language is commonly present in the environment) and EFL (when the language is foreign, and therefore cannot be heard often outside the classroom). This is a fairly substantial difference, as the students, the resources, and even the status of teachers may be quite different in one context or another.

With regard to different geographical settings, a variable that the authors have sometimes heard of in informal conversations or even in some hands-on professional presentations, but not as yet in many scholarly research reports, are the possible differences among NNSs based on their country of origin, or rather the region of the world they come from or the first language they speak (cf. Moussu Reference Moussu2006, for example). An example of that would be a recurrent commentary made by experienced teacher trainers comparing some specific features attributed to Asian NNS teachers to others attributed to European NNS teachers. One word of caution is required here. Even though this is a difference that deserves further research, we must pay attention to Holliday's (Reference Holliday2005, Reference Holliday2008) warning regarding the fact that ‘regional’ or ‘national’ characterization may be another expression of ‘culturism’, as it often comes together with the simplification of ‘the other’, thus simply reinforcing previously acquired unfounded stereotypes.

Another difference that must be necessarily contemplated involves the level at which instructors are teaching: whether they are working in primary, secondary or tertiary education is going to impact on how they conduct their professional activity, what recognition they obtain from it, and their ultimate status within the profession. This variable has already been considered by Llurda & Huguet (Reference Llurda and Huguet2003) in their study on primary and secondary teachers' self-perceptions on their own level of proficiency in English, their teaching ideologies, and ‘political issues’ regarding the role of NNS teachers and English as an international language, but more studies looking at how teachers at different levels teach and how they conceptualize their own teaching are needed.

Yet another variable that was briefly singled out by Medgyes (Reference Medgyes1994) was the amount of time spent in countries in which English is spoken. Medgyes obtained significant correlations between this variable and preference for NS teachers. However, Llurda (forthcoming) found that EFL Catalan teachers with prolonged stays in English-speaking showed a greater awareness of NNS strength and the importance of promoting English as an international language. In other words, they appeared to be less affected by the ‘inferiority complex’ that is much more apparent in some of their colleagues with lesser experience abroad. That study showed, among other things, how important it is to conduct replication studies in order to obtain further evidence to support or discard initial findings.

Another different, though related, factor must be considered: the difference that may exist between NNS teachers who hold degrees in a country that is internationally acknowledged as a home country for that language (in the case of English, those countries would clearly be the UK, the US, Canada, and Australia), and those who do not. In one of the interviews reported by Holliday (Reference Holliday2005), an informant claims that many employers in non-English speaking countries value the possession of certificates from dominant English speaking countries more than any other professional experience. No empirical study has been conducted on this particular difference, but the research question is clearly presented here for any willing researcher to tackle it.

Finally, we must deal with differences among NNS teachers created by different levels in target language proficiency. This is the only difference that has been regularly contemplated as part of the discussion on the role of NNSs in language teaching, and more often than not it has been mentioned as the main handicap suffered by NNS teachers in the exercise of their profession, except maybe for the ‘fortunate ones’ who are sometimes labelled as ‘near-natives’ (J. Liu Reference Liu and Braine1999a; Moussu Reference Moussu2006). Despite Canagarajah's (Reference Canagarajah and Braine1999) strong defence of the teaching potential of NNS teachers, including those whose language skills are low, it is normally assumed that the greater the language proficiency, the better the teacher. However, language proficiency has never been used as an independent variable in order to observe and describe differences among NNS teachers. Such an analysis might greatly contribute to our understanding of the role of the teacher's language proficiency in language teaching.

4.2.2 Research on student teacher supervisors and program administrators

To this day, very little has been done to involve student teachers' supervisors and IEP administrators in research projects. Pasternak & Bailey (Reference Pasternak, Bailey and Kamhi-Stein2004) discussed issues of preparation and professionalism, and Brady & Gulikers (Reference Brady, Gulikers and Kamhi-Stein2004) presented their experience with NSs and NNSs and suggested how TESOL MA programs could cater to NNES student teachers' needs. However, few scholars besides Llurda (Reference Llurda and Llurda2005b) and Nemtchinova (Reference Nemtchinova2005) have surveyed teacher educators regarding the NNS and NS student teachers under their supervision. An increase in research on teacher educators' perspectives on NNSs (as opposed to the student teachers' perspectives) is needed to help NNS student teachers be better prepared to teach in different contexts.

With regard to language program administrators, Mahboob (Reference Mahboob2003) and Moussu (Reference Moussu2006) surveyed several IEP administrators in the United States, but the EFL context is in great need of further studies about practices and beliefs of language program administrators. Individual anecdotes and a few articles (see Braine Reference Braine1998, for example) tell of the difficulties NNS teachers encounter when trying to find teaching positions in an EFL context, and Flynn & Gulikers (Reference Flynn and Gulikers2001) wrote an article that provided several suggestions to NNSs on how to prepare themselves professionally, from their IEP administrators' perspectives. However, issues related to the preparation, hiring, and supervision of NNSs are still rarely, if ever, mentioned in language program administrator handbooks (see Christison & Stoller Reference Christison and Stoller1997, for example).

Further investigations of hiring practices, personal beliefs, and administrative procedures of language program administrators, as well as the study of economical or political factors that could influence hiring and teaching conventions and measures could prove valuable.

4.2.3 More classroom observation based studies

As has been made clear in the above sections, most research on NNSs has heavily relied on teacher self accounts, by means of narratives, surveys, or interviews. Therefore, it is the teachers' voices that are normally heard, and through their perspective a research corpus has been established. In section 4.1.5 above, we have referred to the few classroom observation studies that (to our current knowledge) have been conducted so far, dealing specifically with NNS teachers. This shortage of studies points to a major lack in research using this method.

Pairing native and non-native teachers has been reported as a good way to complement the strengths of both native and non-native teachers (Medgyes Reference Medgyes1994; de Oliveira Reference de Oliveira and Richardson2001, Reference de Oliveira, Richardson and Kamhi-Stein2004). However, this practice has revealed some problems in context like Japan, where this practice has been institutionalized (Kachi & Lee Reference Kachi and Lee2001), and more research is needed. Pairing NSs and NNSs in the same classroom and observing how they distribute their roles in different contexts would certainly help understanding to what degree the performances attributed to each group do actually appear in the classroom.

A further challenge for the future is to set an agenda that seeks to establish connections between what the teachers have reported about themselves and what they actually do in the classroom. This research agenda will need to consider the differences among NNSs, and therefore will have to look for some patterns of generalization without losing track of the essentially individual nature of what may be called ‘the art of language teaching’. We would not be surprised to find out that some generalizations that have been made to all NNSs apply to some but not to others. The route is open and unexplored and the vastness of the task may appear discouraging, but once we have started looking at the nature and characteristics of NNS teachers, it would not be rational to be satisfied with a list of generalizations that supposedly apply to all NNSs, and then leave this field without accomplishing the ultimate goal of describing the nature of language teachers as a way of: (1) removing discrimination against them, and (2) better understanding language teaching in general, and more particularly in those situations in which the teacher is a member of the silent non-native speaking majority in the profession.

4.2.4 Exploration of new methods and topics

The amount of research on non-native English-speaking teachers conducted in the last ten years is impressive, considering the little – and rather scattered – work that had been done in this area prior to 1999. Given the enormous task of presenting the foundations of a new area of research, it is remarkable how much has been achieved in such a short period of time. However, as often happens, recent research has opened new questions and areas of investigation and has necessitated new approaches in future studies. Clearly, more qualitative ethnographic research and the exploration of experimental designs are needed. Longitudinal studies are also required, as very few studies (e.g., Moussu Reference Moussu2006) have looked at their subjects over a period of time. Also, the use of mixed methods and triangulation, such as that used by Morita (Reference Morita2004) and that proposed in Cots, Llurda & Irún (Reference Cots, Llurda, Irún, Camps and Milian2008) combining qualitative and quantitative methods, could provide greater insights into the area. Another aspect that could be considered is students' actual performance, namely, how different teachers' practices affect students' learning.

Some of the questions that have already been answered in exploratory studies, but still deserve further exploration, range from issues of professional self-esteem, collaboration between native and non-native teachers, interaction patterns found in classes with NNS teachers, or even students' judgments of what it takes for a teacher to be a NS or a NNS. Another set of topics that need further attention concerns aspects that have already been researched, but never from a NNS teacher's perspective. One such topic is the effect of teacher education on actual teaching: How different contents and methodologies used in different teacher training programs worldwide affect native and non-native teachers in different settings. Teacher education and its effects on actual teaching practices is a fundamental area of research in order to improve the quality of language teaching, but it will not be complete without a specific observation into the processes by which NNSs gain teaching expertise (Tsui Reference Tsui2003), together with an understanding of how NNS teachers' self-perceptions are altered by experience and expertise.

As stated above, one of the most urgent needs at this point is to develop a research agenda that focuses specifically on classroom observation and the analysis of the actual teaching performance of NNS teachers, complemented by a triangulation of results obtained through different methods. Only by doing this will the field be able to overcome the feeling of stagnation that may eventually appear once the results of recurrent surveys conducted in different contexts and with different subjects fail to add any new perspective on what has already been found. This methodological shift towards classroom practices will surely bring a new set of topics or a new perspective to those already existing, such as the comparison of teachers at different educational levels and in different contexts, or the grammatical knowledge and level of language awareness displayed by NNS teachers in the classroom.