English as a Lingua Franca: Perspectives from International MA TESOL Students in the United States

English learning and teaching have taken many forms in the last years. It is studied as a foreign language and a second language, and with specific purposes in technical fields of knowledge. However, in a broader context, English has become the lingua franca of communication and business. This new paradigm has impacted how teachers and learners see the acquisition of English. While some people prefer native or native-like English learning models, others are getting away from it. That is, some prefer intelligible communication rather than native-likeness. Therefore, this quantitative non-experimental study examined international MA TESOL students' perspectives about their awareness of English as a lingua franca and investigated their aspirations and preferences as English learners. The sample was composed of 20 non-native English teachers. Data were collected through an online survey that examined teachers' preferred models of English, beliefs about the most conducive environments for learning English, aspirations for correctness, intelligibility, and fluency. The findings provided empirical evidence that non-native English teachers are aware of the different English varieties and that they recognize these varieties as valid forms of English. This study found that participants put a higher emphasis on intelligibility than on grammatical accuracy if they thought that certain utterances would not impede communication. However, the results revealed a dual orientation in participants' aspirations, where their strong preferences paradoxically contradicted their awareness of the diversity of English varieties and their emphasis on intelligibility for native-like models of pronunciation and lexical knowledge. The implications of this study suggest that TESOL teacher education programs should provide spaces for analysis of the role of ELF in international learning milieus.


Introduction
English has become the world's lingua franca. English is the chosen language for a wide variety of settings where a neutral language is required. For example, it is used in international transportation, business, the media, the Internet, entertainment, science, and academia. This fact represents one of the significant challenges in the ELT field [1,2]. Various researchers [3][4][5], claim that a change from TEFL (Teaching English as a Foreign Language) or TESOL (Teaching English to Speakers of Other Languages) to a broader platform of TELF (Teaching English as a Lingua Franca) has occurred, especially in academic-related contexts where English has become the means of instruction.
However, one wonders whether or not English teachers are aware of that and prepared to face this new paradigm. Mostly, international TESOL students who are pursuing master's degrees in the United States. That is, we have observed that mater's programs in TESOL in the United States tend to promote more native and native-like varieties of English and teaching models. However, as the existing literature indicates, English as a lingua franca or English as an international language is a reality. This fact has made some English learners and teachers try not to achieve native or native-like language competencies, but to achieve English proficiency skills that allow them to communicate with other English speakers worldwide regardless if their ways of speaking English is considered non-standard by some interlocutors.

Purpose of the Study
Therefore, the purpose of quantitative non-experimental study was to examine if non-native English teachers were aware of the existence of ELF-based teaching practices and to examine their aspirations and preferences for themselves and their students to face the TELF paradigm. The following research questions guided this study.


What variety of English (e.g., British, American, non-native) do NNET aspire to for themselves?  What variety of English (e.g., British, American, non-native) do NNET aspire to for their students?  What environments (native vs. non-native) do they consider to be most conducive for the acquisition of English?  Do they emphasize on accuracy vs. intelligibility?  Is there a relationship between their aspirations and their preferences for accuracy and acceptability?

Literature Review
This next section addresses a literature review about attitudes and perceptions that students have about English as a lingua franca. It also talks about English teaching practices as a lingua franca.

Students' Attitudes and Perceptions towards EFL
Research shows that English learners' attitudes and perceptions towards ELF vary. Sometimes ELF is seen as a viable option for English learners over a native English learning paradigm; however, the opposite occurs. For instance, in an investigation of 127 non-native English-speaking participants from 22 different European L1s, [6] found no significant difference between the perceptions learners have to be exposed to native or non-native speakers as for intelligibility (p = 0.17); however, 79.53% of informants demonstrated a preference for a native-like speaking manner. In addition, 80.16% of participants said that European ELF should be the English variety taught in European schools rather than English as a native language. Cogo [7] found that international English learners indicated a preference for ELF communication skills rather than correctness. However, the findings of perceptions of different accents were somewhat ambiguous since the participants showed a preference for native English accents, but also an interest in non-native accents.
Additionally, in a study that included 69 undergraduate English learners, Xu and Van de Poel [8] discovered that English is recognized as a highly essential means for international communication rather than for connecting socially with native speakers. Also, the study revealed that at a micro level, participants demonstrated a preference for linguistic accuracy measured against the native speaker standard form. Furthermore, in a study with 239 Hungarian university students, Csizér and Kontra [9] found that English as a native language was preferred. However, when it had to do with communication, ELF was accepted. Lastly, Ranta [10] found that English teachers recognized that ELF is sometimes more accepted in the European context as a desired variety of English to learn because it is more likely that learners may end up interacting more with non-native speakers than with native speakers.

Practices in Teaching English as a Lingua Franca
This section focuses on studies that have examined English teachers' attitudes and perceptions of ELF. Dewey [11] conducted a study with ten experienced English teachers enrolled in a DELTA course. Eight of the participants were native speakers; one came from the Outer Circle (L1 Malayan) and one from the Expanding Circle (L1 Czech). The study explored the degree of ELF awareness within this group of teachers. Parts of the themes that were examined were, for example, standard English, English as a global language, English as a lingua franca, the English varieties teachers should be aware of, and which English variety should be presented in the language classroom as a model. Other aspects that were investigated included correctness, acceptability and intelligibility for international communication, and importance for classroom correction. Dewey [11] found that ELF is accepted.
Nevertheless, participants stated that the concept of standard English was limited to only one variety, British English. Pursuing this further, some participants believed that adopting varieties of English, such as Singaporean or Malaysian English as models for instruction, could lead to an interruption in communication between those using these varieties and international interlocutors. As a mechanism to overcome communication breakdowns, participants suggested the adoption of standard British English as the chosen model for teaching in the classroom. It was also found that the different varieties of English should be kept as separate entities, such as avoiding the use of American and British features in combination. In a later study, Dewey [4] examined the relationship between ELF research and current beliefs and practices among English language teachers. The results displayed a noticeable growth in the teachers' awareness of the ELF paradigm and how it fits under the umbrella term "World Englishes." Dewey [4] reported that the majority of participants (only two provided no answer) were able to conceptualize the terms "Global English", "World Englishes," and "English as a lingua franca." The relation of these terms with teaching practices also displayed a more sophisticated understating of their implications for the participants' teaching practices. Further, they showed awareness of English nativization within each community in which it was spoken and the resulting indigenized English varieties that have emerged. These participants were also aware of English speakers' role in the outer and expanding circles and the diversity and plurality showed by the globalization of English. As for the implications of teaching ELF, the participants responded that ELF as a teaching paradigm was acceptable in theory; however, it was also perceived as somewhat unrealistic due to the practical implications of displaying such varieties in the classroom. They perceived these forms to be highly intelligible but also incorrect when comparing them with the established norms. The next section elucidates the methodology used to conduct this study.

Research Design
This next section covers the research methodology and design employed to carry out this inquiry.

Research Approach
This quantitative non-experimental research study used an online survey to investigate MA TESOL graduate students' perceptions about the English varieties they preferred for themselves and their students and their beliefs about the best environments to learn English. Additionally, teachers' opinions and beliefs were correlated with their judgments on the correctness, acceptability, and intelligibility of selected statements from ELF corpora as they were utilized in Dewey [11].

Research Site and Participants
The research site was the Linguistics Department at a Midwestern university in the USA. It was chosen purposefully. The accessible population was (N=30) MA TESOL students, non-native English teachers (NNET). A sample of 20 participants was selected randomly. Participants were informed that their participation was voluntary. The participants' ages ranged between 23 and 53 years, and the mean age was 30; furthermore, 11 participants were female and 9 were male. All participants indicated that they had English teaching experience in international contexts. The least experienced, 38.9%, had taught English for about three years, and the others, 61.1%, had taught from 4 to 7 years. They belonged to 12 different countries, namely, Argentina (1), China (3), Colombia (2), Congo (1), Iraq (1), Libya (1), Nicaragua (3), Oman (1), Saudi Arabia (4), South Korea (1), Vietnam (1), and Yemen (1).


What variety of English (e.g., British, American, non-native) do NNET aspire to for themselves?  What variety of English (e.g., British, American, non-native) do NNET aspire to for their students?  What environments (native vs. non-native) do they consider to be most conducive for the acquisition of English?  Do they emphasize on accuracy vs. intelligibility?  Is there a relationship between their aspirations and their preferences for accuracy and acceptability?

Data Collection
The data were collected in view of several general areas of interest, namely, teachers' preferred models of English, beliefs about the most conducive environments of learning English, aspirations for correctness, intelligibility, and fluency. Each of these constructs was measured by 19 Likert scale statements survey and a judgment task used in Dewey's [11] study. The Likert scale items were formulated as statements, and participants were asked to express their level of agreement. The Likert scale consisted of 5 levels, namely, 1 = Strongly Disagree, 2 = Disagree, 3 = Somewhat agree, 4 = Agree, and 5 = Strongly Agree. Three TESOL and Applied Linguistics specialists established the validity of the survey. They critically examined and judged the questions, and changes were made based on their suggestions. After data collection, all Likert scale questions were analyzed through Cronbach's alpha for internal consistency. In addition, the instrument included Dewey's [11] judgment task where participants were asked to evaluate seven sentences for correctness, acceptability, and intelligibility for international communication, and importance for classroom correction. Figure 1 illustrates one of these items.

Data Analysis
Data analysis involved calculating descriptive statistics, including frequencies and measures of central tendency. Frequency tables were created with percentages of responses. Three dependent t-tests and correlation analyses were conducted to establish the relationship between the variables of interest. The data were analyzed using the IBM SPSS 19 Statistics Package. In the next section, the main findings are presented and discussed.

Main Findings of the Study
This next section covers the main findings of this inquiry. This study sought to explore if NNET were aware of the existence of ELF-based teaching practices and their aspirations and preferences for themselves and their students to face the TELF paradigm. This study used a survey to collect data to inform the proposed research questions.

NNET'S Aspirations towards English
The first six statements in the survey investigated teachers' aspired English models. Each statement was measured on a scale of 1 = strongly disagree and 5 = strongly agree. Participants' responses to each statement were analyzed through descriptive statistics, including frequencies and central tendency measures. Table 1 presents the frequency distributions. For the ease of describing trends towards native or non-native English varieties, the percentages of responses in the categories strongly agree and agree and strongly disagree and disagree have been collapsed. Findings indicated that sometimes participants showed a preference for a native-speaker model, whereas other times, they preferred a non-native speaker model. This alternation of meaning is dependent on how each survey item has been stated. The somewhat category is retained as a midpoint, which shows hesitation between native and non-native preferences.  Table 1 shows that 50% of participants agreed that being perceived as a native English speaker was not a priory but to be intelligible to be understood (S1). The other 20% disagreed with it; whereas, 30% showed a certain degree of hesitation (somewhat agree). As for statement two (S2), 70% of them strongly disagreed regarding their preference for speaking to native speakers than non-native speakers. Only 15% of them preferred to talk to native speakers rather than non-native speakers. Concerning statement three (S3), 50% of respondents demonstrated disagreement, 15% of them agreed, and 30% hesitated by choosing somewhat agree. The distribution of the answers for the fourth statement (S4) showed that the majority of the participants (65%) disagreed with it, 20% of them answered either strongly agree or agree, and 15.0% showed hesitation. Statement five (S5) had the highest Mean (4.2) among all statements. Thus, 85.0% of the answers exhibited a higher agreement; whereas, 10% of the participants disagreed with this statement.
The major trends of participants' aspirations towards native or non-native models of English are best observed through the Mean scores for each of the above described five statements. These means are displayed in Figure 2 as the aspirations towards non-native and native models of English are color-coded in blue and red, respectively. Figure 2 illustrates that the first four statements in the survey revealed that the majority of the informants showed a preference for a non-native vs. a native model of English. These findings are supported by the relatively high mean score (3.6) of Statement 1, showing that as in general, participants were aspiring towards intelligibility vs. native-speaker accents for themselves as speakers of English. The next three statements, 2, 3, and 4, elicited rather low mean scores (below 2.5), revealing a disagreement rather than agreement with these statements. That is, participants' responses to these statements did not demonstrate a preference towards a native model of English. Even though they did not prefer native speakers as conversation partners, they did not indicate that they understood non-native speakers better than native speakers. Further, and they did not perceive a strong non-native accent as negative. However, statement 5 elicited a definite inclination towards a native speaker English model. The mean score of 4.20 states that participants admired non-native speakers who can speak English like educated native speakers.

Aspirations towards a Native or Non-Native Model of English for Students
Statements six to ten (S6-S10) elicited participants' aspirations towards a native or a non-native model of English for their students. The data were analyzed through frequencies and central tendency measures. The frequency distributions are presented in Table 2. For ease of identifying trends, the percentages of responses of the categories strongly agree and agree and strongly disagree and disagree are collapsed in the narrative summary of these results. Thus, statement six (S6) revealed a stronger trend towards a native model of English than towards a non-native model, as 47.3% of the participants agreed, 15.6% somewhat agreed, and 36.9% disagreed. However, an opposite trend, in favor of a non-native model, was observed in response to statement seven (S7), where most of the participants (60%) agreed with it, 10% somewhat agreed, and 30% disagreed. Regarding statement eight (S8), the trend was reversed in favor of native-like proficiency, as 85% of the participants agreed with the statement, 10% showed slight agreement, and only 5% disagreed. For statements nine and ten (S9 & S10), the trend shifted again in favor of a non-native model. Explicitly, for statement nine (S9), about which 60% of the participants agreed, 10% showed a slight agreement, and 30% disagreed. Lastly, as for statement ten (S10), findings showed that 50% of participants agreed, 25% somewhat agreed, and 25% disagreed.
The major trends of participants' aspirations towards native or non-native English varieties for their students are better observed through the Mean scores of the statements mentioned above. The means presented in Figure 3 as preferences towards a native-like model are coded in blue, and those towards a non-native model are coded in red.  , and a belief that students should be exposed to different varieties of English (Mean 3.35). Table 3 displays the frequency distributions of participants' responses concerning statements eleven through sixteen (S11-S16). For statement eleven (S11), the tendency was slightly in favor of English speaking contexts as 36.8% of the participants agreed, 36.8% slightly agreed, and 26.3% disagreed. However, regarding statement twelve (S12), most of them (70%) viewed non-native environments as conducive for acquiring a high proficiency in English, 20% somewhat agreed, and only 10% disagreed. When asked about the global status of English (S13), 52.6% of participants agreed, 31.6% selected somewhat agree, and 15.8% of the participants disagreed. As for statement fourteen (S14), 57.9% of respondents agreed that people of different parts of the world would speak their English variety, 31.6% of them somewhat agreed, and only 10.5% disagreed with this statement. Statements fifteen and sixteen (S15 & S16) measured participants' preferences for native or non-native teachers of English. The overall trends were in favor of native English teachers. Thus, for statement fifteen (S15), 45% of informants agreed, 25% somewhat agreed, and 15% disagreed. A similar tendency was observed in the frequencies for statement sixteen (S16), where 55% of the participants disagreed, 35% somewhat agreed, and only 15% agreed.   Figure 4 illustrates the Mean scores for each of the six statements. Views in favor of native environments and norms are coded in blue (S11 & S15), whereas views about non-native environments and norms are coded in red (S12, S13, S14, & S16).

Views about Native and Non-Native Environments and Norms
As can be seen in Figure 4, respondents considered native English speaking environments and norms to be more conducive for learning English. These findings were supported by the mean scores for statements elven and fifteen (S11 & S15), which were above 3, and by the fact that the mean score for statement sixteen (S16) was below 2.5. On the other hand, participants' responses to statements twelve, thirteen, and fourteen (S12, S13, & S14), revealed their recognition of the status of English as a global language and the fading importance of native-speaker norms.

Views of the Importance of Accuracy Versus Intelligibility
Three statements (S17, S18, & S19) of the Likert scale instrument measured participants' views on the importance of accuracy vs. intelligibility. Table 4 states the frequencies distributions of the elicited responses. Concerning statement seventeen (S17), most of the respondents (52.6%) showed that they were not bothered by students' grammar mistakes in speaking, 21% indicated a slight agreement, and 26.3% expressed disagreement. The responses for statement eighteen (S18) indicated that 35% of the informants considered more important the organization of ideas than grammatical accuracy, 40% agreed to a certain extent, and 25% viewed grammatical accuracy more important than the organization of ideas. About statement nineteen (S19), 55% of the participants gave more weight to fluency than grammatical accuracy when assessing students' oral performance, 30% slightly agreed, and 15% acknowledged grammatical accuracy as more important than fluency in assessing students' oral performance.
The overall trends are better illustrated by the Mean scores displayed in Figure 5. Accuracy was measured through statement seventeen (S17) (in blue), whereas intelligibility (in red) through statements eighteen and nineteen (S18 & S19). 3.40 1.09

Figure 5. Participants' preferences for accuracy and intelligibility
The major trends displayed in Figure 5 revealed consistent preferences for fluency over language accuracy, both in assessing students' speaking and writing performance. Emphasis on fluency vs. accuracy in speaking is shown through the mean scores for statements nineteen and seventeen (S19 & S17). For statement nineteen (S19), the mean score of 3.4 shows a greater concern about fluency than about accuracy. Conversely, the rather low mean score (2.74) for statement seventeen (S17) suggests that the participants were not attributing great importance to grammatical errors in students' speaking performance. A slightly higher emphasis on fluency vs. grammatical accuracy was also observed concerning assessing students' writing performance. The mean score of 3.15 about statement eighteen (S18) shows that participants gave slightly higher weight on the organization of ideas than on grammatical accuracy.

Judgments of Accuracy, Acceptability, and Intelligibility
The last part of the instrument was a replication of the one used in Dewey's study (2011). Specifically, participants were given seven sentences with embedded grammatical errors and asked to judge them given the following four categories: 1) Correctness, 2) Acceptability for International Communication, 3) Intelligibility for International Communication, and 4) Importance for Classroom Correction. These categories were measured on a six-point scale (1-6), with 1 showing disagreement with or lack of a specific characteristic/quality and 6 showing agreement with or high presence of a particular characteristic/quality. Table 6 displays the Mean scores for these utterances.
As illustrated by Table 5, the lowest mean scores were found regarding language correctness, with the scores for the seven utterances ranging between 2.20 and 3.40, and a mean correctness score of 2.96. Among the seven sentences, the highest score was observed about utterance 2, Last summer I was happy because I finally took (got) my driving license, whereas utterance 3, I enjoying listening (to) classical music, received the lowest correctness score. Moreover, it seems that grammatical errors were judged more harshly than lexical errors. For example, utterances 2 and 7, which had embedded lexical errors (took instead of got; draw the limits instead of draw the limit), received the highest mean scores, whereas the reaming five, which contained grammatical errors, received lower mean scores.
Given the following criterion, acceptability for international communication, the mean scores were much higher than those for correctness with means ranging from 4 to 4.45, and a mean acceptability score of 4.33. Utterance 2, Last summer I was happy because I finally took (got) my driving license, was judged as the most acceptable among the seven utterances. It was the same utterance that received the highest score for correctness. Consistent with the correctness judgments, the lowest acceptability score was observed for the same utterance (# 3), I enjoying listening (to) classical music. Among the four judgment categories, the highest scores were obtained in view of intelligibility for international communication, with means ranging between 4.30 and 5, and an overall mean score of 4.76. Utterance 5, My sister has (the) same problem as me, received the highest intelligibility score of 5. Utterance 3, I enjoying listening (to) classical music, was judged the most harshly again as it received the lowest intelligibility score.
In the last category, importance to correct, the means ranged between 3.45 and 4.45, with an overall mean score of 3.88. The highest urge to correct was observed in relation to utterance 3, I enjoying listening (to) classical music. The lowest need to correct was expressed concerning utterance 1, We need to discuss about the problem. Figure 6 displays the main trends found through the measures of central tendency. These trends show that despite the low correctness scores, participants judged the utterances as intelligible and acceptable for international communication. They expressed a moderate urge to correct embedded errors.
In the next part of the analysis, participants' judgments of correctness were compared with their judgments of acceptability, intelligibility, and importance to correct through 3 dependent t-tests. This analysis sought to examine whether the judgments of correctness were statistically different compared to the judgments on the remaining criteria. Table 6 displays the results for the t-test analysis.  Note: * = significant at alpha .05: ** = significant at alpha = ,001 The first comparison between correctness and acceptability showed a statistically significant difference, t(19) = .-2,125, p = .047, effect size d = .47. In other words, participants' correctness scores were significantly lower than their acceptability scores. The next comparison between correctness and intelligibility was highly significant, t(19) = .-5.56, p <.001, effect size d = 1.24 and showed that correctness judgments were significantly lower than intelligibility judgments. The third comparison between correctness and importance to correct showed no significant differences, t(19) = .-1.74, p = .097, effect size d = .38. Although the mean correctness score was lower than the mean importance-to-correct score, this difference was not large enough to reach statistical significance. Among the three comparisons, the greatest difference was observed between correctness judgments and intelligibility judgments. Following Cohen's (2008) reference values of .2, .5, and .8 for small, medium and large effects, respectively, it can be concluded that the difference in judgments was not only statistically significant but had practical importance since it revealed a high effect size value of 1.24.
This section presented a detailed account of the analyses of the data elicited through a Likert scale survey and Dewey's [11] judgment task. The results showed interesting and sometimes opposing trends between participants' preferences, perceptions, and judgments. The interpretation and discussion of these trends are presented next.

Discussion of Main Findings
This next section provides a discussion of the main trends found in the participants' perceptions of and attitudes towards experiences in ELF about the theoretical and empirical literature. They will be discussed in terms of: 1) Teachers' preferences for a model of English for themselves and their students, 2) Teachers' beliefs about the most conducive environments for English acquisition, 3) Teachers' views on the importance of accuracy versus intelligibility, and 4) Teachers' judgments of accuracy, acceptability, and intelligibility.

Preferences for a Model of English for Themselves and Their Students
When showing their preferences regarding non-native or native varieties of English, findings displayed two divergent trends. On the one hand, the results showed that the participants did not prefer to be perceived as native English speakers. They showed a preference for interacting with non-native English speakers rather than with native speakers. As well, informants were aware of the existence of different varieties of English and their role in international communication and were more receptive to non-native accents. Their preferences for non-native norms were also apparent in their interest in making their students aware of multiple English varieties from the Outer and Expanding Circles [12] and promoting communication skills rather than native-like competence. On the other hand, the participants favored native norms of proficiency in the areas of pronunciation, lexicon, and idiomatic expressions. Furthermore, they perceived native speakers to be more intelligible, and most of them (85%) reported that they admired non-native speakers who have developed native-like English speaking skills. Likewise, 85% of the participants declared that they helped their students develop English idiomatic and lexical knowledge and 63.1% showed a preference for their students to achieve native-like pronunciation.
These findings were similar to those found in other investigations where English teachers were aware of the importance of non-native English forms. However, participants were also attached to the native norms, especially about pronunciation. Jenkins [13] reported that English teachers have a desire to achieve native-like pronunciation as one of their markers of being recognized and recognizing themselves as successful English speakers and teachers. This duality can also be explained in Stanojević and Smojver's [14] terms of liberal and traditional attitudes towards English. The participants of this study projected themselves as more liberal English users, accepted non-native use of English as valid, and desired native use of English neither for themselves nor for their students. Most probably being themselves second-language speakers of English, they were aware of the difficulties of developing a native-like ability in a language, and were, thus, more open-minded about different non-native varieties of English. This group of English teachers was also exposed to various non-native forms of English because they were all master's degree students in a TESOL program with a higher percentage of international students than American students. Thus, their perceptions about language features that make an intelligible speaker were based on the more global use of English and its importance in terms of international communication.
On the other hand, these teachers were more traditional when attempting to make their students more intelligible and believed that the native norm would provide them and their students the appropriate patterns for being intelligible. The notion of intelligibility was also found in the lexical domain, where participants perceived teaching idiomatic expressions in the classroom to be important because of their role in the pragmatics of social interactions with native speakers.

Beliefs about the Most Conducive Environments for English Acquisition
The dual orientation towards native and non-native norms was also noticed in the participants' beliefs about whether or not native environments were more conducive to English acquisition. The answers to this construct followed the same abovementioned patterns where the preferences towards native norms were constrained to specific domains. Nonetheless, overall, a strong preference for non-native norms was apparent. Upon being asked for what they thought were the contexts most conducive for English acquisition, participants preferred English speaking countries. Contradictorily, they considered it to be possible to acquire a high level of English competence in learning contexts outside native speaking countries.
Participants' responses showed this divergent trend concerning the global status of English and its multiple uses worldwide. Participants recognized the global expansion of English and the existence of different varieties of native and non-native Englishes. This strong acceptance (89.5%) of emerging English varieties was evident in participants' beliefs that in the future, people from diverse parts of the world would speak their chosen variety of English. Paradoxically, though, when asked about their preferences about native or non-native English teachers, only 15% of the participants preferred English classes taught by non-native teachers. A dual orientation was also found by Galloway [15], who reported that despite their awareness of the global status of English and the diversity of non-native varieties, young Japanese English learners preferred native English teachers. Their preference was assumed to be the result of their interest in idiomatic lexicon acquisition and a personal desire to achieve a native-like pronunciation, areas in which native English teachers were considered more skillful.

Views of the Importance of Accuracy Versus Intelligibility
Overall, the primary trend in terms of participants' preferences for accuracy or intelligibility showed that they preferred their students to be intelligible more than being grammatically accurate. If the primary purpose of using a language is to communicate as noted by Widdowson [16], the participants' emphasis on intelligibility vs. accuracy can be attributed to their adherence to the view that the attainment of communicative competence in a second language is more important than the attainment of linguistic accuracy. This finding suggests that the participants were influenced by the principles of the Communicative Approach [17], which emphasizes language functions over forms [18]. Their preference for fluency and their acceptance of grammatical errors can be seen in the ELF perspective as metaphors of an evolutionary trend in favor of communication [19]. However, these teachers did not wholly separate the linguistic forms from their functions. They saw the organization of ideas and grammar accuracy in their students' oral presentations as equally important.

Judgments of Accuracy, Acceptability, and Intelligibility
When judging the seven ELF sentences in terms of accuracy, acceptability, intelligibility, and the importance of classroom correction, the participants evaluated these sentences as highly intelligible and very acceptable for international communication. Nevertheless, the scores for correctness (M=2.96) were statistically lower when they were compared with the scores for intelligibility and acceptance for international communication. This finding suggests that despite their realization of the grammatical incorrectness of the given utterances, the participants recognized that the errors would not be detrimental to communication. Moreover, when scores for correctness and importance to correct were examined statistically, they revealed a lack of significant difference. That is, although the mean correctness score was lower than the mean importance to-correct score, this difference was not large enough to reach statistical significance.
What this finding implies is that the participating teachers' gave priority to intelligibility vs. accuracy even though they realized that the given utterances had errors. Likewise, their urge to correct the errors in classroom contexts was moderate rather than high. The observed trends are similar to the ones found by Dewey [11] when applying this same task to a group of English teachers (8 native and 2 non-native). Dewey [11] discovered that English teachers found his ELF corpus very intelligible and very acceptable for international communication even though his participants assigned lower scores for correctness.
The trends described by both groups indicated a pattern of teachers valuing a promotion of communicative competence. Dewey [11] proposed that to achieve a certain level of communication in ELF interaction, speakers have to accommodate their production to continue the interaction. The accommodation processes between speakers make it possible for them to understand a particular utterance despite its form not being fully linguistically competent. The acceptance of the nonstandard forms to communicate with others in ELF situations was explained by Friedrich[20] to be a result of a desire to convey meaning in interlocutions to reach an agreement between the speaker and his/her interlocutor, agreements that in ELF situations are motivated by a desire to be successful, for example, at business, Internet growing, or academic development.
However, in addition to the general trends observed in the total mean scores, including all seven utterances, additional insight was brought by the descriptive statistics for each specific utterance. For example, concerning correctness, utterances 2 and 7, which had embedded lexical errors (took instead of got; draw the limits instead of draw the limit) received the highest mean scores, whereas the remaining five utterances that contained grammatical errors received lower mean scores. Among them utterance # 3, I enjoying listening (to) classical music, received the lowest mean score. Consistent with the correctness judgments, this same utterance received the lowest acceptability and intelligibility ratings and the highest score for importance to correct (Mean = 4.45). Thus, a more detailed look at the participants' judgments of acceptability, intelligibility, correctness, and importance to correct, show a link between their perceptions of the severity of the errors and the implications for intelligibility. Based on their judgments of the severity of the errors, the teachers formed their judgments about the need for remedial classroom work.
In sum, the participants showed a strong adherence to the principle of intelligibility vs. accuracy, yet it should be noted here that this preference was also dependent on their judgments of the severity of the errors. Unfortunately, this emerging trend was based on limited qualitative analysis, but it provides interesting suggestions for future research and should be pursued in a much more comprehensive and systematic way.

Conclusions
Findings concluded that the research participants were aware of the different English varieties. They acknowledged that these varieties as valid forms of English. Moreover, participants seemed to put a higher emphasis on intelligibility than on grammatical accuracy if they thought that certain utterances would not impede international communication. However, the results also revealed a dual orientation in participants' aspirations, where their strong preferences paradoxically contradicted their awareness of the diversity of English varieties and their emphasis on intelligibility for native-like models of pronunciation and lexical knowledge. This contradiction between awareness and aspirations is not new. It has been documented in the findings of other studies [11,14,15,20], and it most probably corresponds to a developmental confusion in the emerging paradigm of ELF. Whether this confusion will be resolved, it is a question of how ELF continues to evolve. We recommend more quantitative and qualitative research related to English as a lingua franca in other contexts with samples that includes new generations of young and adult learners who are learning or are planning on learning English in contexts where English is not the main means of communication. Future studies should include samples of future English teachers both native and non-native speakers to examine their English learning and teaching preferences.