Navigation – Plan du site

AccueilNumérosVol. 38 N°2ArticlesTraining for English-Medium Instr...


Training for English-Medium Instruction in French Higher Education: Keeping Pronunciation in Focus

Formation à l’EMI dans l’enseignement supérieur en France : focalisation sur la prononciation
Alice Henderson


L’internationalisation de l’enseignement supérieur français s’accompagne souvent d’enseignements disciplinaires dispensés en anglais (ou English-Medium Instruction, EMI). La formation proposée aux enseignants de disciplines non-linguistiques qui se lancent dans ce défi devrait inclure un travail tant sur la langue que sur les contenus, ainsi que sur les spécificités pédagogiques qui apparaissent lorsqu’on propose des contenus par le biais d'une langue étrangère. Plus particulièrement, je défends l’idée qu’un travail explicite sur la prononciation doit figurer dans cette formation, en visant plus que l’intelligibilité. Une formation à l’anglais oral (pour les enseignants non-linguistes comme pour les étudiants, en tant que locuteurs et interlocuteurs à tour de rôle), doit viser à améliorer deux aspects : l’attitude face à la variation de la langue orale et une plus grande conscience de l’impact d’un parler non-standard, appelé ici foreign accented speech (FAS). FAS se définit comme tout parler ayant des traits acoustiques qui relèvent de la/des langue(s) maternelle(s) du locuteur. Dans cette optique, je présente d’abord la prononciation en tant que matière dans les formations d’enseignants d’anglais du monde anglophone (ESL et EFL) et la prononciation d’une langue non-maternelle comme un champ de recherche international. Ensuite, je décris brièvement la formation à l’EMI en France, avant de présenter quelques notions pertinentes provenant des recherches sur l’impact social et cognitif du FAS. Les adaptations souhaitables chez les locuteurs et interlocuteurs sont expliquées, puisque l’objectif est non seulement de modifier leurs attitudes envers la variation de la langue orale mais également d’améliorer leur aisance à déchiffrer le FAS. Cela m’amène à décrire quelques exercices de formation, avant de conclure en préconisant un rôle plus important pour les enseignants d’anglais dans l’internationalisation de l’enseignement supérieur français.

Haut de page

Texte intégral


1This paper is based on my keynote address at the APLIUT conference in 2018, the title of which echoed Gilbert’s article published in 2010, “Pronunciation as orphan: what can be done?”. Gilbert was referring to EFL/ESL training programmes in the English-speaking world and how they failed to teach the what, how and why of pronunciation teaching. In French higher education, there is a lack of training programmes that specifically prepare teachers to teach their subject matter in English: the what, how and why of combining language and content. In general, good quality training and support for EMI teachers includes the pedagogical specificities of combining the two in one teaching context, in addition to:

the involvement of higher education language specialists and their close collaboration with content specialists, an appropriate CLIL methodology in light of the increased cognitive load, the appropriate level of vehicular language competence for both teachers and students (for the latter, an absolute minimum of B1+ on the Common European Framework of Reference for Languages, CEFRL). At stake is a decrease in the quality of both teaching and learning — the very opposite of the spirit of the Bologna process. (Taillefer 35, citing Bartik et al. 2009).

  • 1 See Derwing and Munro (“Pronunciation Fundamentals”) for an excellent discussion of the construct o (...)

2I will argue that work on pronunciation – from the perspective of both speakers and listeners – must not only be included in such training, but it must go beyond improving intelligibility. Beyond the continuing terminological debate about which features influence intelligibility the most1, training in the specificities of spoken English – for both EMI teachers and students, as interchangeable speakers and listeners – should also aim to improve their reactions to foreign-accented speech (FAS), defined here as any speech which exhibits acoustic features from a speaker’s mother tongue(s).

3Other papers in this issue deal with the specificities of EMI within French higher education and the broader issue of internationalisation. Therefore, I will start by briefly describing pronunciation as a subject in English language teacher training and pronunciation as a research field in the English-speaking world. This is followed by a description of EMI training in French higher education: training for teachers who would like to teach their subject in English, as distinct from training for English language teachers. Then, I present a few key findings of research into the social and cognitive impact of FAS, before outlining what is involved for both listeners and speakers, if we want to modify their attitudes to spoken language variation and to improve their cognitive comfort in processing FAS. I conclude by suggesting exercises for training and, in line with Taillefer, by advocating that language teachers play a role within the internationalisation of French higher education.

1. Pronunciation as orphan?

4Gilbert (“Pronunciation”) summarised research showing that training to teach pronunciation was absent from EFL/ESL teacher training programmes in Britain, Canada, the UK and Australia and that research was lacking and/or not easily accessible to teachers. A European-wide study of English pronunciation teaching found similar results (Henderson et al.). On a more positive note, Levis (“Cinderella”) shows that since 2010, the field of second language pronunciation has become much more visible, with over 40 professional books, numerous special issues of international journals and the first professional journal devoted to the subject, the Journal of Second Language Pronunciation:

We have dedicated conferences and webinars and electronic conference proceedings. We have greater visibility in journals. We have software, but we need more. The infrastructure of the field is taking shape, but we need one more thing, teacher education. We know that most teachers, whether native or non-native, feel unconfident about how to teach pronunciation, what to teach, … (Levis “Cinderella” 12).

5There is thus a considerable body of research and a structured field for English language teachers around the world to tap into, but adequate training in how to teach second language pronunciation remains a weak point in language teacher education, according to Levis (“Cinderella”).

6Turning to French higher education, training in the pedagogy of English-medium instruction (EMI) is an area of teacher education where progress is also needed. I will argue in this paper that pronunciation training must be included in such training. Moreover, it seems logical to call French universities’ highly trained English language teachers into action, especially those from the LANSOD sector (Languages for specialists of other domains), to support their non-linguistic colleagues who would like to teach in English. Doing so would allow French higher education to put more effective EMI into place more rapidly, by avoiding the phase of “pronunciation as orphan” that continues to characterize EFL/ESL teacher training programmes in the English-speaking world.

  • 2 Défi international « Dispositif d'accompagnement pédagogique et linguistique pour l'enseignement en (...)

7While research into EMI is widespread, studies of EMI in French universities are not (Pagèze and Lasagabaster). Moreover, according to Pagèze and Lasagabaster, it is rare for French universities to provide any support or training for teachers wishing to teach their subject in English, and what little exists is eclectic. One example of excellence is the 8-strand programme put in place by Pagèze and her colleagues at the University of Bordeaux2, with a set of modules focusing on both pedagogical skills and language skills – including a module on pronunciation focused on intelligibility. The course is aimed at practising teachers of non-linguistic subjects and students who have already finished their undergraduate degree. A post-training survey of participants showed that the training had modified their view of EMI, “moving them on from their initial concern about their language skills, to the classroom situation of EMI and being able to identify aspects of language that support classroom communication.” (Pagèze and Lasagabaster 302). The change applies to both French and English, as

90% of respondents said that following the training they had changed their teaching approach in French and/or English. All respondents reported feeling more confident in teaching their discipline through English (Pagèze and Lasagabaster 303).

8Teacher confidence is an example of the psychological factors, both affective and cognitive, that affect professional performance and cannot be disregarded in an EMI context. One major affective factor is the close link between pronunciation and identity. Teachers teaching their subject in English may be non-native yet quite proficient users of English; nonetheless they put their professional competence on the line, in public, when they use English to work with a body of international students. One unsettling question many such teachers ask themselves is “will students take me seriously if I speak this way?”, and they may also experience discomfort or embarrassment at adopting new prosodic patterns or articulatory settings.

9In cognitive terms, the demands on teachers and learners are even greater in an EMI context. Teachers should already master English to a high level; C1 or C2 on the CEFR scale is required in French secondary schools to teach a subject in English. However, evidence gained from (informally) observing 12 of our colleague from non-linguistic subject undergoing EMI training at the Université Savoie-Mont-Blanc confirms that not all such university teachers have this level of mastery. These 12 teachers also admitted that teaching in English tired them mentally and physically, more than teaching in French, their native language, and similar observations have been made by teachers at the University of Applied Sciences in Vienna (Richter). For learners, the theory of cognitive load posits that the load involved in learning content is higher when that content is presented in another language (Roussel & Tricot). And while cognitive load related to processing content via another language can be mediated to some extent by teacher traits or pedagogical techniques (Roussel & Tricot), it remains an important factor to consider when implementing an EMI programme. One example of the potential impact comes from a study by Piesche, et al., who had expected deeper processing of science content by pupils in a bilingual German-English classroom condition. However, their results showed that monolingually educated pupils outperformed the bilingual pupils. The question becomes: what could be done to lessen cognitive load for learners?

10One option would be to include learners in EMI training, instead of focusing solely on teachers. Gilbert (“Pronunciation”) concluded her article by describing ways to boost the skills and confidence of teachers, and Levis (“Cinderella”) concludes his overview by calling for “increased access to teacher training, another part of our disciplinary infrastructure” (12). Colleagues from across France have confirmed that EMI initiatives focus solely on teachers. However, given the two-way nature of spoken interaction, both teachers and students are important to the solution, especially as European EMI spreads to the undergraduate level and to a wide variety of contexts. As Pagèze and Lasagabaster argue (307), we need to move away “from framing EMI as a language problem to framing it as a specific disciplinary communication context”, which requires us to alter our views of the role of language in learning and teaching. Just as importantly, the teachers who went through their programme in Bordeaux “also started to think of English as a lingua franca, which helped to dispel some of their fears as non-native speakers.” (Pagèze and Lasagabaster 308). Acknowledging this role for English is a first step toward recognizing the role of pronunciation as a social marker. It is thus crucial to adopt a holistic, socially-situated view of language use in the EMI context; failure to do so could have serious ramifications, as the next section shows.

2. Realities of Foreign-Accented Speech

11The language which comes out of our mouths is one of the most salient social markers (Giles) and therefore is a powerful vector for discrimination and bias, whether applied to someone speaking our mother tongue with an “accent” or someone speaking another language. Even though most people would overtly subscribe to tolerance, subconscious language-based discrimination, in the form of latent perceptual stereotypes and prejudice, is a well-documented phenomenon. Gluszek and Dovidio survey over four decades of research in social psychology, showing that more negative perceptions are attributed to speakers with non-native accents, and the stronger the accent the more negative the evaluations. This section argues that, because FAS is widespread within higher education, it is crucial to understand unconscious reactions to it and the potential impact on behaviour. In other words, the impact of subconscious bias on learners’ cognitive comfort must not be ignored when implementing EMI. To this end, this paper draws on the notions of cognitive load and processing fluency because, as Fraser states:

The goodwill of local people must go beyond mere tolerance and patience, to active understanding of, and accommodation to, the linguistic needs of the other (e.g. to respond in a way that is perceived as encouraging by the recipient, not merely intended as such by the speaker). (Fraser A118).

12Understanding this will help us to better prepare students and teachers for successful, professional communication and learning. The American Psychological Association defines learning as:

  • 3

the acquisition of novel information, behaviors, or abilities after practice, observation, or other experiences, as evidenced by change in behavior, knowledge, or brain function. Learning involves consciously or nonconsciously attending to relevant aspects of incoming information, mentally organising the information into a coherent cognitive representation, and integrating it with relevant existing knowledge activated from long-term memory3.

  • 4 As defined on the web site of the American Psychological Association: (...)

13Good teaching should facilitate the necessary attending, organizing and integrating, ideally with the lowest possible cognitive workload, defined by the APA as “the relative demand imposed by a particular task, in terms of mental resources required”4. Numerous teaching techniques, such as scaffolding, can reduce this workload, whereas learning via a non-mother tongue increases it (Roussel et Tricot). In EMI, processing information via speech often means via foreign-accented speech and therefore, heightened cognitive load.

14Research into cognitive load raises the issue of whether it is justifiable to try to teach content through a non-native language (Roussel et al.). It has not been shown that such learning is impossible, nor that it is always more difficult or easier, nor even that certain features in FAS have greater or lesser degrees of impact (e.g. non-standard syntax, non-native prosody). Nevertheless, accent does appear to have an impact on learning. For example, in their study of the effects of speaker variability on EFL learners, Gao et al. found that proficient English learners learn well despite encountering numerous accents (the multiple-speaker condition), whereas less-proficient learners benefit from dealing with only one accent (the single-speaker condition). Hence, accent, via its cognitive impact, is a crucial variable in EMI.

  • 5 It is interesting that more than a century ago Sweet had already proposed a distinction between int (...)

15Furthermore, socio-cognitive research has shown that FAS is evaluated less positively (e.g. Gluszek and Dovidio; Beinhoff) and that both explicit and implicit attitudes are activated in Implicit Association Tests (e.g. Pantos and Perkins). In numerous studies exploring language-based discrimination, speakers with an accent received lower status ratings (Miller and Hewgill; Brennan and Brennan; Bresnahan et al.; Lindemann; Lev-Ari and Keysar). Studies of lecture-style speech have shown similar results (Rubin & Smith; Kavas & Kavas). Particularly relevant to an EMI context is the fact that few adults acquire native or native-like pronunciation in languages beyond their mother tongue (Flege et al.). However, the good news is that intelligibility and accentedness are partially independent variables (Munro and Derwing), as shown in multiple studies of the correlations between accentedness, comprehensibility and intelligibility5. Work by Derwing, Munro and others over the last twenty years has been central in clarifying these three constructs and their interactions, shifting the emphasis in L2 pronunciation instruction from the nativeness principle onto intelligibility (see Derwing and Munro “Training”; “Pronunciation”).

16It is, however, necessary to go beyond these constructs, and the concept of cognitive load, to the socio-psychological concept of processing fluency, which is defined as the way speech stimuli are judged as a function of the cognitive load involved in their processing (see Reber and Schwarz; Oppenheimer). For example, when listeners are distracted by a speaker’s accent while they are processing information, cognitive resources are drained by the effort to decode speech. This was dubbed the “vampire effect” by Mai and Hoffman (149). Moreover, intensive cognitive processing can lead us to judge a person less favourably, if we find it difficult to decipher their speech.

17Studies of judgments of competence reveal the force of another factor, the native-speaker model. In Volin et al. study using Czech, French and American students as listener-raters, a preference for native speakers over FAS speakers was statistically highly significant. However, the French students showed a strong out-group preference and gave very low ratings to speakers of English who had a strong French accent. In contrast, the Czech raters showed a weak in-group preference (as would be indicated by a preference for Czech-accented speech) and were less critical of English spoken with a French accent. Rubin and Smith and Kavas and Kavas explored reactions to the academic discourse of university professors who were non-native speakers of English. Their most striking result is that even when rater-participants were warned about the tendency toward bias, they still rated the FAS speakers as less competent. Hence, the raters’ social evaluations were inseparable from their perception of speech. Furthermore, Berkovcová et al. showed that FAS speakers are rated less favourably, even when rater-listeners do not understand the speaker’s language. Their listener-raters were twenty university students of Spanish and twenty students unfamiliar with Spanish, and only the rhythmic patterns of Spanish speakers were modified to create the stimuli. Again, the stimuli presenting a modified rhythm were rated lower. These studies all confirm that FAS powerfully influences people’s behaviour – even if on a subconscious level and in reaction to a single, prosodic modification.

  • 6 See also Hanzlikova and Skarnitzl for a replication of Lev-Ari and Keysar’s study.

18To conclude, intelligibility is not the sole concept that must be taken into account when studying reactions to FAS. Speech perceived as non-native impacts negatively on credibility judgments, and this holds for both native and non-native listeners6. Because the credibility of teachers may be questioned solely on the basis of their speech, it is disrespectful to throw them to the lions without somehow addressing this issue. The next two sections provide suggestions for how training can improve listeners’ (i.e. students’) and speakers’ (i.e. teachers’) attitudes to variation and their ability to comfortably process FAS.

3. Improving attitudes to variation

3.1. Training Listeners

At the University of New England, NSW, Australia, Fraser developed a programme to train students and staff from English-speaking and non-English-speaking backgrounds (ESB and NESB) in both speaking and listening. Analysis of the survey results and follow-up interviews reveals a mixture of positive and negative comments about NESB lecturers – but also about the lack of patience shown by ESB students. Two-way feedback workshops were organized in order to counter the fact that “despite goodwill, each ‘side’ harbours considerable misunderstanding of the needs, motivations and intentions of the other” (A116). The ESB and NESB students worked on developing what he termed their willingness to communicate (Macintyre) and their ability to cope with the pauses, false starts and misunderstandings, which typically pepper spoken interactions. Fraser’s study debunked the very common myth held by NESB students, according to which speaking fast shows fluent mastery of the language:

Very good results were gained from workshopping their own experience of struggling to keep up with English speakers and encouraging them to speak for their ESB interlocutors as they would wish to be spoken to, that is, pausing frequently, monitoring facial expressions to be sure of understanding, and allowing space for questions and clarification. It was also noted that pausing was not just good for the listener but allowed the NESB student time to prepare the next phrase, resulting in lowered stress all round (Fraser A118).

19Fraser concluded that “developing the intercultural speaking and listening skills of local people, as well as being valuable in its own right, can offer a cost-effective way to help improve international students’ English” (Fraser A118). Although Chinese students studying in English in Australia are not in the same situation as students (from France and/or other countries) attending a class taught in English at a French university, the conclusion is just as valid: communication is a two-way street and any training for EMI must involve both students and teachers.

20In another example of research into improving attitudes to variation in the context of spoken interaction, Derwing et al. helped Anglophone social workers in Canada to improve their understanding of Vietnamese-accented English, as well as their attitude toward working with the Vietnamese community. The social workers received eight weeks of intercultural and/or linguistic training. While the results did not show any significant improvement in listening comprehension, those who had received training (either uniquely intercultural or intercultural coupled with linguistic training) did show more empathy and confidence. Those who

received explicit instruction regarding the characteristics of Vietnamese-accented English showed significantly greater improvement in confidence that they could interact successfully with individuals who speak English as a second language, while the group that received only cross-cultural awareness instruction showed moderate gains. Similarly, the accent-trained group believed that their ability to understand foreign accents improved as a result of instruction to a significantly greater degree than the other groups (Derwing et al. 245).

21Improving the social workers’ confidence in their ability to successfully interact undoubtedly has an impact on their willingness to do so and may even encourage them to actively seek out such contact, thus providing them with opportunities to successfully understand speakers, in what becomes a virtuous circle.

22Perspective-taking exercises are another potentially useful means of raising awareness of attitudes and their impact on our thinking and behaviour. An example exercise to use with students would be to discuss the case in figure 1 below.

Figure 1: Example exercise to improve listeners’ attitudes to variation

Two students have just walked out of the classroom. They are criticizing the teacher’s accent in quite a harsh and negative way.

*(for language majors):

Describe how the phonological traits which the two students are discussing are realised in a native variety of English and how these differ from the teacher’s realisations.

*(for language and non-language majors): If you were that teacher, what would you do differently

- in order to have a more native-like accent?

- in order to be more easily understood?

23To the extent that attitudes and beliefs underlie our behaviour, such exercises could be useful in an EMI training programme.

3.2. Training Speakers

24Small, regular exchange groups are perhaps the most obvious means of modifying speakers’ attitudes toward variation, for instance via sharing EMI experiences or having meta-discussions of the concepts “native-like” and “intelligible”. Although not all European universities have a wide variety of language and cultural backgrounds in their teaching staff, these exchanges can change people’s perspectives, as in the project at the University of Bordeaux. This is also where virtual exchange could play a role, allowing educational staff from around the world to participate in discussions. One such example is the ERASMUS+ EVOLVE project, which seeks to promote virtual exchange, defined as:

a practice, supported by research, that consists of sustained, technology-enabled, people-to-people education programmes or activities in which constructive communication and interaction takes place between individuals or groups who are geographically separated and/or from different cultural backgrounds, with the support of educators or facilitators7.

25Virtual exchange has promise because it combines the potentially powerful impact of intercultural dialogue with the ease-of-access afforded by digital technologies. It is easy to imagine teachers in several different countries “micro-teaching” each other and sharing feedback. Inevitably there will be differences in pedagogical skills, language skills and accentedness. However, personal experience of others’ EMI teaching – both good and bad – may help boost teachers’ confidence, by providing points of comparison. Although the goal is to modify teaching practices, as Pagèze and Lasagabaster found, other important changes usually follow:

Not only did they change their teaching practices by making their classes more student-centred, but they also started to think of English as a lingua franca, which helped to dispel some of their fears as nonnative speakers. Changing perspectives on the way language, communication and thus learning can be managed in the university teaching classroom allows teachers to feel more legitimate in teaching in English because it lowers the stakes for teachers – language load is shared and managed in a learning approach based on co-construction of knowledge. This is helpful because it shifts teacher identity from being a model of linguistic perfection to a facilitator and manager of a classroom situation and places the student at the center of learning. (Pagèze and Lasagabaster 308)

26Learner-centred teaching can thus contribute to teachers’ feelings of legitimacy, a crucial factor in successful EMI.

4. Improving cognitive ease/comfort

4.1. Code vs. signal

27Cognitive processing of speech is a vast field of research where at least one fundamental distinction, between code (which is language specific) and signal (which is not) (Bradlow), is crucial to the present paper. The relevant acoustic signal here is the speech stream, the acoustic wave, travelling from a speaker’s mouth to a listener’s ears. The code is the constructed, language-specific system or the filter in the brain which decodes speech we hear or, in the other direction, encodes our ideas and intended meanings when we want to speak. For example, when I began learning Polish, if a native speaker of Polish repeatedly said “no, no” in a conversation, I decoded it as a negative, whereas a native speaker of Polish would interpret it entirely differently. “No” is a positive discourse marker in Polish, showing engagement with the speaker’s topic and encouraging them to continue. In the case of FAS, when the signal reaches our ears and then our brain, we tend to apply our native language ‘filter’ to decoding the signal (Troubetzkoy). Conversely, when we want to communicate and exchange with speakers of other languages, we could try to modify the way we encode our ideas, to make it easier for them to grasp our meaning; this takes conscious, attentive effort. Modifying the signal, in contrast, is often quite instinctive – we speak more loudly as a bus goes past or close the window to block out noise.

28To conclude, improving cognitive comfort or ease can involve the code and/or the signal, where adjusting the code is more of a conscious challenge. For people involved in EMI in French higher education, if they have no auditory deficiency, the real difficulty is at the decoding and encoding levels, in other words the differences between French and English.

4.2. Training Listeners

  • 8 It is interesting to note that French pupils have improved their listening comprehension skills by (...)

29One of the main challenges in French EMI remains the level of English proficiency of local French students. EMI is often portrayed as the solution to improving their proficiency, by virtue of mixing with international students and being exposed to the language. However, French students will need additional support if their listening comprehension8 is to be sufficient:

France (along with Spain and Turkey) is one of the few countries where foreign students’ English-language proficiency is rated higher than that of local students. Data from the 2012 First European Survey of Language Competences indicates that just over 20% of French school leavers attain the B1- B2 bands of language competence in English (Pagèze and Lasagabaster 292).

  • 9 See Taylor and Hayes-Harb for an overview.

30Given the mismatch between English spelling and pronunciation, a logical conclusion is that learners need to learn two sets of vocabulary, one for the written form of words and another for what Cauldwell calls the “sound shapes” of words. At the Université Savoie-Mont Blanc on-line exercises were developed to raise awareness of such shapes. FAS adds another layer of complexity to the task of decoding the stream of speech, and some research-based, technological solutions are appearing to address this. One example is High Variability Phonetic Training, which offers systematic variation training in the perception of speech sounds9. The hypothesis underlying HVPT is that being exposed to systematic variability of vowel and consonant contrasts makes it easier for people to perceive – and later to produce them. Testing different facets of this hypothesis has important implications for EMI: could exposure to systematic variability facilitate the generalisation of adaptation to FAS? In fact, such adaptation can occur with less than one minute of exposure to FAS, or even within two to four sentence-length utterances (Clarke and Garret). Furthermore, after exposure to multiple talkers of a particular accent, individuals showed talker-independent adaptation to FAS (Bradlow and Bent). In another study, Baese-Berk et al. asked the next logical question: what types of exposure would help listeners with novel speakers of FAS? The study used monolingual English listeners but widened the types of exposure, bringing in speakers from five distinct language backgrounds. Their listeners were able to generalise to novel speakers and accents. These findings support the idea that such generalisation is due to exposure during training to systematic variation in FAS. While the latter study did not identify the similarities across the different speakers, “there are some global properties that are similar across many foreign-accented talkers of English (e.g., difficulties with the crowded vowel space, diminished unstressed vowel reduction, and reduced speech rate)” (Ibid.179). These three properties alone constitute a firm starting point for any training that seeks to prepare listeners and speakers for EMI.

4.3. Training Speakers

31Helping teachers with pedagogical design is fortunately a key component of many EMI training programmes and workshops. More specifically, training should raise awareness of cognitive load and ways to manage it (Sweller; Sweller et al.). Any pronunciation work with speakers should focus on those features of English which make it easier for their listeners to understand their ideas, even when those ideas are expressed via FAS.

32Thomson’s decades of work on HVPT resulted in the 2012 launch of his website10 where, free of charge, people can train “the brain to recognise new sounds – an essential foundation for improved pronunciation”. The website includes those vowel contrasts which have proved difficult for many learners. Some of these contrasts are essential to what Bradlow refers to as clear speech: “a distinct speaking style that talkers adopt when the listener has speech perception difficulties due to a hearing loss, background noise, or a different native language11”. Clear speech is oriented toward the Other and involves improving access to the signal (e.g. via repetition), as well as making more apparent the most salient language-specific features of the code (e.g. via paraphrasing or reformulation). Regarding English, Bradlow’s numerous studies of clear speech (e.g. Smiljanić and Bradlow) recommend that speakers make the following adjustments to acoustic and phonetic features:

  • slow down;

  • pause at meaningful points;

  • clearly distinguish between long and short vowels, especially in stressed syllables;

  • enhance distinction between minimally different words (e.g. pass/path, import/export);

  • vary pitch range to signal meaning and to avoid monotony.

33Bradlow’s work on adaptation to variants of a native language laid the groundwork for further studies, such as Wright et al., which posit that a general learning mechanism may be involved in both speech processing and language acquisition. This paper is interesting here because of its implications for training both speakers and listeners, or EMI teachers and students. Wright et al. ran two experiments involving: 1) acquisition of a novel phonetic category (specifically, a new Voice Onset Time) and 2) adaptation to FAS. Both experiments alternated periods of practice with periods of stimulus exposure without practice. This mirrors language acquisition of our mother tongue or any further language, because it involves periods when we speak it and listen to it – we attend to it consciously – and periods where we are immersed in the language but do not consciously attend to it. The authors give the example of cooking while listening to a podcast or radio broadcast in a non-native language. Their findings are intriguing:

These practice and exposure periods yielded little to no improvement separately, but alternation between them generated as much or more improvement as did practicing during every period. Practice appears to serve as a catalyst that enables stimulus exposures encountered both during and outside of the practice periods to contribute to quite distinct cases of speech learning. It follows that practice-plus-exposure combinations may tap a general learning mechanism that facilitates language acquisition and speech processing” (Wright et al. 928).

  • 12 It is worth remembering that, in order to attract foreign students, France does not need to offer c (...)

34One of the key reasons so many French universities are eagerly promoting EMI as part of their internationalisation strategy is to ‘kill two birds with one stone’; that is, they believe EMI will attract foreign students and benefit home students12. It is assumed that simply by being in a class where a non-native language is spoken will improve students’ skills in that language. This is not always the case, because listening to a language spoken with an accent may drain students’ cognitive resources (the previously mentioned “vampire effect”, Mai & Hoffman); in other words, students will not necessarily improve their mastery of a language simply because they are taking a class in that language. Nonetheless, Wright et al.’s “practice-plus-exposure combinations” could constitute a useful guiding principle when developing training courses for teachers and students alike. For example, the following exercises have worked in my classrooms. In all cases, they are minor adjustments to exercises with which many language teachers are familiar.

35Examples of practice-plus-exposure exercises with students:

- 10-15 minutes of weekly homework doing online “noticing” exercises focused on salient features (e.g. phrasal stress, Hahn; ‘sound shapes’); there are no right or wrong answers, merely awareness raising questions which are discussed in class;

36- a video marking the 400th anniversary of Shakespeare’s death13 provides excellent modeling of phrasal stress, around the well-known “to be or not to be” line. Learners can watch the video to notice this feature, then realise it aloud – chorally and individually. Then they can identify it in novel situations and/or create their own.

37Examples of practice-plus-exposure exercises with teachers:

38- drawing on Gilbert’s Prosody Pyramid (Gilbert “Teaching”), exercises with breath groups can be created. A simple start is to ask people to say, in at least two different ways let’s eat grandma, or a woman without her man is lost. The follow-up discussion is a good time to establish some shared meta-language about how pausing and intonation contribute to meaning.

39- accent extension (Giles et al.) can be achieved in only six sessions and with speakers of any level by working on articulatory setting for 5-10 minutes every week (Ashton and Shepherd; Honikman; Messum and Young). Despite any initial reticence learners might have regarding the physicality of articulatory setting work – which may seem out of synch with their notions about what language classes should involve – such exercises can profoundly change their understanding of how speech is produced and categorised.

Conclusion and future perspectives

40I started this paper by describing pronunciation as both a subject in English language teacher training in anglophone countries and as an international research field, and this led into a brief look at EMI training in French higher education. I then explained why it is important to improve both the skills and confidence of EMI teachers and students. In particular, the latter should be trained to cope with a variety of speakers and accents, on both cognitive and affective levels. This could alleviate situations where, tired of trying to decipher a teacher’s accent, they experience frustration and give up, putting the blame on the speaker. Moreover, instead of putting the blame solely in one camp, boosting speaker and listener intercultural speaking and listening skills might even prove to be more cost-effective (Fraser A118).

41English language teachers could also be involved in supporting their non-linguist colleagues and their students in their institutions. They should keep in mind the fact that spoken interaction involves speakers and listeners, both of whom can adapt their attitudes and behaviours. Given the existence of relevant research and professional events as well as the core of professional English language teachers in place, English language teachers – especially those in the LANSOD sector already familiar with the interplay between language and non-linguistic content – are essential players in effective EMI for French higher education. Teacher partnerships should be set up, in which an English language teacher and a non-linguist teacher are officially paired and exchange regularly. Simply having a regular time and space in which to exchange experiences can be hugely beneficial for both. Such exchanges can be virtual and even asynchronous.

  • 14 This includes native and non-native English users, those who have passed a competitive national exa (...)

42EMI therefore provides an opportunity for teachers of language and non-language subjects to collaborate and refresh their pedagogy. As foreign-accented English is so widespread in EMI and can have a powerful subconscious impact on speakers and listeners, pronunciation remains the heart of many issues which need to be addressed: the what, how and why. What pronunciation work can be done in short EMI training courses, for both speaking and listening goals? How will it be done, using which models and focusing on which aspects of spoken English? And why will certain features be given priority over others? Regarding these issues, English teachers can more effectively support their EMI colleagues if they themselves feel sufficiently trained in teaching pronunciation. Whether or not native and non-native English language teachers in France14 have received extensive theoretical training in phonetics and phonology is independent of how confident they feel about teaching pronunciation to students (Henderson et al.), let alone to fellow colleagues. Gilbert (“Teaching”) suggests a tool to help boost ESL/EFL teachers’ confidence in their ability to teach pronunciation, in addition to improving their teaching skills. Her Prosody Pyramid provides “a simple framework to understand how pronunciation is a coherent system, not just a list of abstractions” (Gilbert “Pronunciation” 4). But English language teachers first need to be aware of such research and remain up-to-date, which is why researchers also have an important role to play, by making relevant research into spoken language accessible. They could even provide practical advice, for example, on how to lower cognitive load or how to mediate language-based discrimination.

43To conclude, EMI is an opportunity to focus on the two-way nature of interaction, as the guiding principle for classroom exchanges, teacher support and training for speakers and listeners. Doing so will bring us one step closer to reducing the impact of our subconscious on our actions, which is a powerful and empowering possibility for all concerned.

Haut de page


Ashton, Helen, and Sarah Shepherd. Work on your Accent. HarperCollins, 2012.

Baese-Berk, Melissa et al. “Accent-Independent Adaptation to FAS.” Journal of the Acoustical Society of America, vol. 133, no. 3, 2013, pp. 174-180.

Beinhoff, Bettina. Perceiving Identity through Accent: Attitudes towards Non-Native Speakers and their Accents in English. Oxford, Peter Lang, 2013.

Berkovcová, Zuzana, et al. “Vliv temporálních manipulací na vnímání kompetence mluvčího.” Studie z aplikované lingvistiky, 7, 2016, pp. 7-19. [Effect of temporal manipulations on the perception of speaker competence].

Beuzon, Sylvie, and Etienne Dalibard. “Compétences en anglais en fin de collège: CEDRE 2004-2010-2016 (Note d’information No. 17.20).” Paris, Ministère de l’Education Nationale DEPP, 2017, p. 4.

Bradlow, Anne. Clear Speech: Applying Research Findings to the Classroom. [Recorded webinar]. Retrieved from, 2007.

Bradlow, Anne, and Teresa Bent. “Perceptual Adaptation to Non-Native Speech.” Cognition, vol. 106, no. 2, 2008, pp. 707-29.

Brennan, Eileen, and John Brennan. “Measurements of Accent and Attitude toward Mexican-American Speech.” Journal of Psycholinguistic Research, vol. 10, 1981, pp. 487-501.

Bresnahan, Mary, et al. “Attitudinal and Affective Response toward Accented English.” Language and Communication, vol. 22, 2002, pp. 171-185.

Cauldwell, Richard. Phonology for Listening: Teaching the Stream of Speech. Birmingham,, 2013.

Clarke, Constance and Merrill F. Garrett. “Rapid Adaptation to Foreign-accented English.” The Journal of the Acoustical Society of America, vol. 116, no. 6, 2004, p. 3647.

Derwing, Tracey, et al. “Teaching Native Speakers to Listen to Foreign-Accented Speech.” Journal of Multilingual and Multicultural Development, vol. 23, no. 4, 2002, pp. 245–259.

Derwing, Tracey, and Murray Munro. “Training Native Speakers to Listen to L2 Speech.” Social Dynamics in Second Language Accent, edited by John Levis and Alena Moyer, Mouton De Gruyter, 2014, pp. 219-236.

----. Pronunciation Fundamentals: Evidence-Based Perspectives for L2 Teaching and Research. Amsterdam, John Benjamins Publishing Co., 2015.

Flege, James, et al. “Factors affecting strength of perceived foreign accent in a second language.” The Journal of the Acoustical Society of America, vol. 97, no. 5, 1995, pp. 3125-3134.

Fraser, Helen. “Speaking and Listening in the Multicultural University: A Reflective Case Study.” Journal of Academic Language and Learning, vol. 5, no. 1, 2011, pp. A110-A128.

Gao, Yuan, et al. “Effects of Speaker Variability on Learning Foreign-Accented English for EFL Learners.” Journal of Educational Psychology, vol. 105, no. 3, 2013, pp. 649-665.

Gilbert, Judy. Teaching English Prosody with the Prosody Pyramid. Cambridge University Press, 2008.

---. “Pronunciation as Orphan: What Can Be Done?” Speak Out, vol. 43, 2010, pp. 3-7.

Giles, Howard. “Teachers’ Attitudes towards Accent Usage and Change.” Educational Review, vol. 24, no. 1, 1971, pp. 11–25.

Giles, Howard, et al. “Speech Markers in Social Interaction.” Social Markers in Speech, edited by Klaus Scherer and Howard Giles, Cambridge University Press, 1979, pp. 343–381.

Gluszek, Agata, and John Dovidio. “The Way They Speak: A Social Psychological Perspective on the Stigma of Nonnative Accents in Communication.” Personality and Social Psychology Review, vol. 14, no. 2, 2010, pp. 214–237.

Hahn, Laura. “Primary Stress and Intelligibility: Research to Motivate the Teaching of Suprasegmentals.” TESOL Quarterly, vol. 38, 2004, pp. 201–223.

Hanzlíková, Dagmar, and Radek Skarnitzl. “Credibility of native and non-native speakers of English revisited: Do non-native listeners feel the same?” Research in Language, vol. 15 no. 3, 2017, pp. 285-298. doi 10.1515/rela-2017-0016.

Henderson, Alice, et. al. “The English Pronunciation Teaching in Europe Survey (EPTiES): Selected results.” Research in Language, vol. 10, no.1, 2012, pp. 5-27. doi 10.2478/v10015-011-0047-4.

Honikman, Beatrice. “Articulatory settings.” In Honour of Daniel Jones, edited by David Abercrombie, et al., Longman, 1964, pp. 73-84.

Kavas, Aysel, and Alican Kavas. “An Exploratory Study of Undergraduate College Students’ Perceptions and Attitudes toward Foreign Accented Faculty.” College Student Journal, vol. 42, 2008, pp. 879-890.

Lev-Ari, Shiri, and Boaz Kayser. “Why Don’t We Believe Non-Native Speakers? The Influence of Accent on Credibility.” Journal of Experimental Social Psychology, vol. 46, 2010, pp. 1093-1096.

Levis, John. Intelligibility, Oral Communication, and the Teaching of Pronunciation. Cambridge University Press, 2018.

---. “Cinderella No More: Leaving Victimhood Behind.” Speak Out, vol. 60, 2019, pp. 7-14.

Lindemann, Stephanie. “Koreans, Chinese or Indians? Attitudes and Ideologies about Nonnative English Speakers in the United States.” Journal of Sociolinguistics, vol. 7, 2003, pp. 348-364.

Macintyre, Peter. “Willingness to Communicate in the Second Language: Understanding the Decision to Speak as a Volitional Process.” Modern Language Journal, vol. 9, no. 4, 2007, pp. 564-576.

Mai, Robert, and Stefan Hoffmann. “Accents in Business Communication: An Integrative Model and Propositions for Future Research.” Journal of Consumer Psychology, vol. 24, no. 1, 2014, pp. 137-158.

Messum, Piers, and Roslyn Young. “Bringing the English Articulatory Setting into the Classroom: (1) The Tongue.” IATEFL PronSIG Journal, vol. 57, 2017, pp. 29-39.

Miller, Gerald, and Murray Hewgill. “The Effect of Variations in Nonfluency on Audience Ratings of Source Credibility.” Quarterly Journal of Speech, vol. 50, no. 1, 1964, pp. 36-44.

Munro, Murray, and Tracey Derwing. “Foreign Accent, Comprehensibility, and Intelligibility in the Speech of Second Language Learners.” Language Learning, vol. 45, 1995, pp. 73-97. doi:10.1111/j.1467-1770.1995.tb00963.x.

Oppenheimer, Daniel. “The Secret Life of Fluency.” Trends in Cognitive Science, vol. 12, 2008, pp. 237-241.

Pagèze, Joanne, and David Lasagabaster. “Teacher Development for Teaching and Learning in English in a French Higher Education Context.” L’Analisi linguistica e letteraria, vol. 25, 2017, pp. 289-310.

Pantos, Andrew, and Andrew Perkins. “Measuring Implicit and Explicit Attitudes toward Foreign Accented Speech.” Journal of Language and Social Psychology, vol. 32, no. 1, 2012, pp. 3-20. doi:10.1177/0261927X12463005.

Piesche, Nicole, et al. “CLIL for all? A randomised controlled field experiment with sixth-grade students on the effects of content and language integrated science learning. ” Learning and Instruction, vol. 44, 2016, pp. 108-116. doi: 10.1016/j.learninstruc.2016.04.001.

Reber, Rolf, and Norbert Schwarz. “Effects of Perceptual Fluency on Judgments of Truth.” Consciousness and Cognition, vol. 8, 1999, pp. 338–342.

Richter, Karin. English-Medium Instruction and Pronunciation: Exposure and Skills Development. Multilingual Matters. 2019.

Roussel, Stephanie, et al. “Learning Subject Content through a Foreign Language Should Not Ignore Human Cognitive Architecture: A Cognitive Load Theory Approach.” Learning and Instruction, vol. 52, 2017, pp. 69-79, doi:10.1016/j.learninstruc.2017.04.007.

Roussel, Stéphanie, and André Tricot. “Emile/CLIL, EMI vs TCC : des espaces de recherche si proches et si lointains.” Alsic [on-line], vol. 20, no. 3, 2017. doi : 10.4000/alsic.3126.

Rubin, Donald, and Kim Smith. “Effects of Accent, Ethnicity, and Lecture Topic on Undergraduates’ Perceptions of Nonnative English-Speaking Teaching Assistants.” International Journal of Intercultural Relations, vol. 14, 1990, pp. 337-353.

Sweet, Henry. The Practical Study of Languages: A Guide for Teachers and Learners. Henry Holt and Co., 1900.

Smiljanić, Rajka, and Ann R. Bradlow. “Temporal organization of English clear and conversational speech.” Journal of the Acoustical Society of America, vol. 124, no. 5, 2008, pp. 3171-3182. doi: 10.1121/1.2990712.

Sweller, John. “Instructional Implications of David C. Geary's Evolutionary Educational Psychology.” Educational Psychologist, vol. 43, no. 4, 2008, pp. 214-216.

Sweller John, et al. Efficiency in Learning: Evidence-Based Guidelines to Manage Cognitive Load. San Francisco, Pfeiffer, 2006.

Taillefer, Gail. “CLIL in Higher Education: The (Perfect?) Crossroads of ESP and Didactic Reflection.” ASp, vol. 63–64, 2013, pp. 31-53.

Thomson, Ron., 2012.

Troubetzkoy, Nikolai Sergeyevich. Principes de Phonologie. Klincksieck, [1939] 1967.

Volin, Jan, et al. “Perceptions of Foreign-accented Lecture-style Speech: Please Take This Seriously.” The Pronunciation of English by Speakers of Other Languages, edited by Jan Volin et al., Cambridge Scholars Press, 2018, pp. 63-80.

Wright, Beverly, et al. “Enhancing Speech Learning by Combining Task Practice with Periods of Stimulus Exposure without Practice.” Journal of the Acoustical Society of America, vol. 138, no. 2, 2015, pp. 928-937. doi: 10.1121/1.4927411.

Haut de page


1 See Derwing and Munro (“Pronunciation Fundamentals”) for an excellent discussion of the construct of intelligibility, and Levis (“Intelligibility”) for the pedagogical implications.

2 Défi international « Dispositif d'accompagnement pédagogique et linguistique pour l'enseignement en contexte international et multiculturel. » See and


4 As defined on the web site of the American Psychological Association:

5 It is interesting that more than a century ago Sweet had already proposed a distinction between intelligibility and native-like performance.

6 See also Hanzlikova and Skarnitzl for a replication of Lev-Ari and Keysar’s study.


8 It is interesting to note that French pupils have improved their listening comprehension skills by 6% over the past ten years, according to recent Education Ministry data (Beuzon and Dalibard). This is thought to be the result of greater availability of subtitled rather than dubbed films and TV series, as well as massive informal audio input.

9 See Taylor and Hayes-Harb for an overview.


11 See also

12 It is worth remembering that, in order to attract foreign students, France does not need to offer courses in English, for it is already “the third country in the world for welcoming international students and French is the third language learnt in the world” (Pagèze and Lasagabaster).


14 This includes native and non-native English users, those who have passed a competitive national exam to become teachers as well as those educated in English-speaking countries – and these categories are not mutually exclusive.

Haut de page

Pour citer cet article

Référence électronique

Alice Henderson, « Training for English-Medium Instruction in French Higher Education: Keeping Pronunciation in Focus »Recherche et pratiques pédagogiques en langues [En ligne], Vol. 38 N°2 | 2019, mis en ligne le 30 juin 2019, consulté le 29 mai 2024. URL : ; DOI :

Haut de page


Alice Henderson

Alice Henderson est maître de conférences en anglais et membre du laboratoire LIDILEM à l'université Grenoble-Alpes, où elle enseigne l'anglais de spécialité aux étudiants de Licence Sciences et Technologies, ainsi qu'aux étudiants du Master en Ingénierie Nucléaire. Pendant 24 ans elle a enseigné à l'université Savoie – Mont Blanc, aux étudiants de LLCE Anglais et du secteur Lansad en sciences humaines et sociales, où elle coordonnait les cours d'anglais. Ses recherches sont centrées sur la prononciation de l'anglais (son enseignement et son apprentissage), la perception d'une variété orale marquée par un accent, English Medium Instruction (EMI), et à présent elle revient aux travaux basés sur les corpus.

Articles du même auteur

Haut de page

Droits d’auteur

Le texte et les autres éléments (illustrations, fichiers annexes importés), sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search