Interview with Prof. Jennifer Jenkins


We were thrilled to have the opportunity to catch up with Professor Jennifer Jenkins at the University of Southampton recently, as she inspired so much of our work at ELFpron. It’s 16 years since Prof. Jenkins published  her research leading to the creation of the Lingua Franca Core (LFC) so we decided it was time to ask her what she is researching now, and how her earlier work on pronunciation helped to inform her current work on multilingualism.

We hope this video might be a useful introduction for the teaching training classroom and anyone interested in better understanding ELF. If you are a fan of video content, check out our new page linking to all the ELFpron training videos, available to watch for free.

Many thanks to Prof. Jenkins for her time and continued support.


ASEAN school exchange


When I heard that a small group of teens from my school in Thailand were going on an ASEAN exchange to Indonesia and wanted help preparing to use English in that context, I jumped at the chance to talk to them about ELF.

What is ASEAN?

ASEAN stands for Association of Southeast Asian Nations. The member states are Indonesia, Thailand, Malaysia, the Philippines, Singapore, Brunei, Cambodia, Laos, Myanmar, and Vietnam. There has been increasing focus on teaching children about ASEAN because the countries have forged closer ties in 2015 with preparations for a common market (the ASEAN Economic Community, or AEC).


What was the aim of the workshop?

To raise students’ awareness of the main differences between their own accents and the accents they might hear in Indonesia, and to build confidence interrupting and asking for clarification in the event of not understanding someone.

What materials were used?

I came across the Tumblr Accent Challenge while hunting for clips on YouTube of Indonesian teenagers speaking English. People from all over the world have uploaded videos of themselves pronouncing a set list of words and answering the following questions:

  • Pronounce the following words: Aunt, Roof, Route, Theater, Iron, Salmon, Caramel, Fire, Water, New Orleans, Pecan, Both, Again, Probably, Alabama, Lawyer, Coupon, Mayonnaise, Pajamas, Caught, Naturally, Aluminium, GIF, Tumblr, Crackerjack, Doorknob, Envelope, GPOY.
  • What is it called when you throw toilet paper on a house?
  • What is a bubbly carbonated drink called?
  • What do you call gym shoes?
  • What do you call your grandparents?
  • What do you call the wheeled contraption in which you carry groceries at the supermarket?
  • What is the thing you change the TV channel with?
  • Choose a book and read a passage from it.
  • Do you think you have an accent?
  • Would you like to be a wizard or a vampire?
  • Do you know anyone on Tumblr in real life?
  • End the audio post by saying any THREE words you want.

I used this particular clip and this handout of the Tumblr Accent Challenge.

How was this used in the workshop?

  1. As you can see, the language level is quite high, so we went through the list of words and questions before listening. It was also useful for the students to try saying the words in the list to prepare for their task while listening, which was to think about the following question: “Do the people in the video say any sounds differently to you?”

  2. After watching the video, the students picked up on some quite specific points, e.g. the use of /d/ in ‘the’,  and other comments were more general like ‘they go up and down a lot’. It was also interesting because the speakers in the video answer the question ‘Do you think you have an accent?’ with ‘No’, which the students thought was funny, because to them the Indonesia teenagers had quite distinctive accents. They said they’d never heard Indonesian accents before.

  3. This was a useful springboard into talking about the students’ own pronunciation because it helped them to appreciate how people in Indonesia might feel about their Thai accents. I gave them a list of pronunciation features from the Lingua Franca Core that often affects the intelligibility of Thai L1 speakers’ English. You can download the handout here and adapt it as necessary. I asked the students to rate how clearly they think they can pronounce each of the features. In doing so, they were saying the words aloud and practicing them at the same time. Then we focused on the areas they felt they wanted more help with. We also talked about the importance of being aware that these features could lead to breakdown in communication, so you can choose different words if it seems that one particular word is not clear to your listener.

  4. Then we worked on interrupting people to say you don’t understand. First, we brainstormed phrases you could use (there’s space to do this at the bottom of the handout). This was unproblematic because knowing what to say is not the problem – actually saying it is another thing! Interrupting someone is culturally quite uncomfortable for Thai people to do, so it took a lot of practice. I spoke too quickly and the nominated student had to interrupt and ask me to speak slower. Then we also tried asking someone to explain one specific word they couldn’t understand. So I deliberately spoke about a topic they understood but inserted one difficult word.

The students responded very enthusiastically and I could see some real ‘lightbulb’ moments. These students were aged 16 and 17, so much older than I currently teach, and therefore able to discuss language variation and issues surrounding accents in a mature way. There were only seven of them and they had been selected to go on the trip. This was quite a different experience to my usual teaching, which is far more limited by time, large numbers of students, and exams. Still, it was an exciting project to be a part of, very briefly, because this is ELF in action. Maybe next time they’ll take me with them… !


How to do a needs analysis with a multilingual class (now with video!)


One of the first posts published by ELFpron was about how to conduct an ELF-informed pronunciation needs analysis with a multilingual class. This environment is ideal for developing ELF intelligibility since the students in a multilingual class are, by definition, using English as their lingua franca.

The activity format is simple: students dictate sentences to each other; you listen and take notes on their pronunciation during the process; and you collect what they wrote to analyse later in more detail (with reference to the Lingua Franca Core) the students’ apparent ability to understand and be understood in an ELF context.

There are step-by-step instructions in the original post, along with examples of what students might produce, so we won’t repeat those here.

But now, we are delighted to be able to share a video of this activity being carried out in a small intermediate-level general English class – and here it is!

Many thanks to the students who volunteered to take part in this demonstration. Your participation is very much appreciated!

Interview with ELTABB


Shortly after our presentation at this year’s IATEFL Conference, we were invited to be interviewed by Dale Coulter of ELTABB, the English Language Teachers’ Association of Berlin-Brandenburg.

In this short interview, Dale asked us to explain the nature of pronunciation in contexts where English is being used as a lingua franca and what this might mean for English language classrooms.

You can watch the interview below, or by clicking here.

Identifying and practising thought groups


This is the second in a pair of posts on the theory and practice of teaching nuclear stress for English as a Lingua Franca (ELF).  The first post explained what nuclear stress is; this post will demonstrate one way of working on this with students.

In the first post, we talked about tone units.  These are short segments of speech which feature one prominent syllable (see the post for a fuller explanation).  They are also sometimes called thought groups.

David Brazil explains in his excellent course on teaching pronunciation that:

“An appreciation that speaking involves one in adding tone unit to tone unit as one proceeds, not, as one tends to think, word to word, is an important part of the awareness on which its successful use depends.” (1994: 3)

The activities suggested in this blogpost take as a starting point this notion of grouping words into units.  The idea is to identify boundaries between groups, rather than identifying features within groups.  I find that this approach is a good way in to working more precisely on nuclear stress, which can be much harder for both teachers and learners to get their heads round!

There are different ways to identify and practise using tone units/thought groups in the classroom.  One way I have tried and found useful is to use general listening practice as a springboard for receptive pronunciation work.

In other words:

  • You do some listening exercises with students as you normally would, focusing on general comprehension and/or comprehension of detailed meaning and/or specific information in the recording.  This is how you would normally listen to something you wanted to hear about.
  • You then might discuss the content of what you heard, again engaging primarily with the message rather than how it was encoded.
  • Then, rather than moving on to something else, you go back to the recording (or a small part of it) and focus intensely on the sound, drawing attention to how the speaker divides up the stream of speech into thought groups.  (I prefer to use the term ‘thought group’ with students as it seems easier for them to grasp and work with than ‘tone units’.)

The idea is to focus first on what is said, and only then how it is said.

Brazil observes that there are two very good reasons for staging pronunciation study in this way:

  1. because “we do not normally attend consciously to the pronunciation of the language we hear or speak.  It consequently requires considerable concentration to do so.  It is better, therefore, if students are not compelled to do it at the same time as they are having to cope with the quite demanding business of putting together or responding to what is being said.  [They should] be thoroughly at home with the content of [a communicative event], so that it has all become as ‘automatic’ as possible before they embark upon the much less natural business of listening for, and reproducing, particular sound patterns.” (Brazil, 1994:4)
  2. because focusing on overall prosody of pronunciation, such as intonation, should focus on “speaking language which is carrying a message, and doing so in some situation in which that message matters to both speaker and listener.  In this way it contrasts, for instance, with the practice of using lists of words to perfect the pronunciation of particular sounds.” (Brazil, 1994:4)

Here is an example of how you might practise identifying tone units with a group of roughly intermediate-level students, using a very short video from the BBC News website.

I chose this specific video for 3 main reasons:

  1. I thought its content would amuse and engage my students.
  2. Its style, speed and length were appropriate for their level and for demonstrating tone units/thought groups very clearly.  This was the first lesson we’d ever worked on this aspect of pronunciation, so I felt that having a reasonably uncontroversial demonstration of how it works was important.
  3. As a news clip, its style and organisation were somewhat clearer and more formal than spontaneous speech.  Roach (2013:367-368) observes that for beginners (which I interpret to include novices to pronunciation analysis in this way, if not actually very low-level learners in general), “it is best to start on slow, careful speech – such as that of newsreaders – before attempting conversational speech.”

Note: the full video is about a minute and a half long.  You only need to use the first 29 seconds.  Here it is:

First: general listening comprehension

  1. Get students to brainstorm unusual sports and competitions.
  2. They watch the video once and identify what unusual sport/competition is featured. (answer: competitive eating)
  3. They watch again and answer some questions: When did the event start? How many hot dogs did he eat?  How long did this take?  Did he feel good about this? (answers: nearly a century ago; 69; 10 minutes; yes, he was full of confidence)
  4. Get them to discuss what they think of this contest.  For example, does it count as a sport?  Is it a good/safe/healthy idea?  Would they try it?  How does a person practise for this sort of event?!  And so on.

Next: identify ‘thought groups’

  1. Give students the tapescript.  It should be typed without any punctuation, so this does not (mis)lead them to assume punctuation = pronunciation.  It should have a few pauses marked already (the convention is to use forward slanting lines), to help get them started.  You can download the one I prepared here.  You’ll see that I’ve included a typed, punctuated tapescript on the other side of the handout for students to look at later and compare/contrast punctuation marks with thought groups divisions.  The part they need to use for the following analysis looks like this: nuclear stress tapescript
  2. Get the students to mark IN PENCIL where they think there is a natural break between words.  They should do this alone, then can compare before you play the recording again.
  3. They listen and compare their notes with when the speaker actually pauses.
  4. Repeat steps 2 and 3 as necessary – it might take about 3 or 4 times before they feel like they’ve caught everything.  (This is why you suggest they use pencil to make their notes!)
  5. Conduct feedback.  This can take some time, as there will undoubtedly be some disagreement and you’ll need to refer back to the recording and discuss with students why they marked pauses in particular places.

On this last point, note that there is not necessarily a strict right/wrong answer key for this activity.  The point is to raise students’ awareness of how speakers break up the stream of speech, and how this facilitates listeners’ understanding.  Brazil points out that:

“when we are using recorded data it is not always possible, even for practised ears, to agree about what is happening.  It is both honest and expedient to admit that there is sometimes room for doubt.  If and when disagreement becomes an issue, it is best to represent it to students as a reason for reassurance: if the experts can’t always agree, there is nothing to worry about if they can’t!” (Brazil, 1994:5-6)

When I did this particular exercise with students, I noticed that they struggled particularly to predict where the speaker would pause when there was a prepositional phrase involved.  They tended to assume the speaker would pause before the noun, not before the preposition, perhaps because they saw the noun as the ‘head’ of the phrase.  For example, they would predict:

he was full of // confidence

instead of :

he was full // of confidence

Similarly, they would predict verbs always falling at the end of a thought group, such as:

and this is // Joey Chestnut

instead of:

and this // is Joey Chestnut

When the difference was pointed out to them and they listened again, they were astounded at how counter-intuitive this felt to them.  It appeared to be quite a common difficulty, regardless of my students’ different first-language backgrounds.

Finally: practise dividing short texts into appropriate ‘thought groups’

  1. Now distribute short texts to students, into which they must insert appropriate points to pause.  The texts should be of a similar length to what they’ve just listened to – it seems very short, but there’s quite a lot of cognitive effort involved and believe me, this will be plenty for them to work with!  They should also not involve too much new vocabulary, as this gets in the way.  For my class, I used this resource from One Stop English, as it was on the topic of competition (like the video clip they’d just watched) and was graded very low, so they could worry about pronunciation and not about unfamiliar lexis.  I adapted it very slightly to suit the number of students in my class and to make their texts a uniform length – you can download my adaptation here.
  2. Once they’ve marked where they think it would be appropriate to pause, they should read their texts to each other.
  3. Their peers should comment on how they thought it sounded – natural? comfortable? appropriate? clear?

With any lesson like this, it’s helpful to finish with some reflection on why this is so useful and how they might apply it in ‘real life’.  For example, when giving a presentation, when dictating information over the phone, etc.  It is important to draw students’ attention to the fact that of course they cannot prepare everything they say every day in this way; but by taking some time for deliberate practice of this area, they build up their awareness of it, with a view to making this part of their active pronunciation repertoire over time.


Brazil, D. (1994). Pronunciation for advanced learners of English (Teacher’s book). Cambridge: Cambridge University Press.

Roach, P. (2013). English phonetics and phonology: A practical course. (enhanced EBook edition)Cambridge: Cambridge University Press.


Understanding nuclear stress


This is the first in a pair of posts (read the second post here) on the theory and practice of teaching nuclear stress for English as a Lingua Franca (ELF).

This post aims to answer the following questions:

  1. What is nuclear stress?  What isn’t it?
  2. How is this relevant to ELF pronunciation?

For one reason or another, English language teachers often don’t meet this terminology–let alone fully understand what it means–until they do a higher-level qualification, such as the Cambridge Delta or Trinity DipTESOL.  The principles behind nuclear stress aren’t really all that tricky; the main stumbling block seems to be how much terminology is involved.  But have no fear; all this will be explained below…

Everything you always wanted to know about nuclear stress* (*but were afraid to ask)

Before we get down and dirty with nuclear stress, let’s start by revising some basics.

English words are made up of syllables.   We won’t go into too much depth here about what precisely constitutes a syllable as it’s somewhat besides the point, but if we can agree for the time being that the word “English” has 2 syllables and the phrase “I speak English” has 4, then that should be enough for us to move on.

Some syllables are stressed.  We can probably agree, for example that the first syllable in “English” is stressed.  Producing a stressed syllable requires more muscular energy.  Recognising a stressed syllable requires us to perceive its prominence (we’ll come back to this in more detail later).  This is what we’re asking our learners to do when we put a new word on the board in class, for example, and ask them “Which syllable is stressed?”

So what makes a syllable prominent?  In my experience, trainee teachers are often taught the simple formula: louder, higher, longer.  This seems to me a reasonable simplifcation.  Not everyone finds it easy to identify stress when asked to – usually, they do it naturally without even thinking about it!  So the ‘louder, higher, longer’ mantra helps make this tricky area more tractable.

Pitch is particularly important.  As Roach neatly summaries (2013:203), “if all syllables [in a word] are said with low pitch except for one said with high pitch […] then the  high-pitched syllable will be heard as stressed and the others as unstressed.  To place some movement of pitch (e.g. rising or falling) on a syllable is even more effective in making it sound prominent.”  And, as Roach succinctly puts it, “the word we use for the overall behaviour of the pitch is tone.” (2013:315)

Here’s are two very simple examples of different tones used to change the intended meaning of the same one-syllable utterance:

Example 1:   Yes.   [falling tone, simply answering a yes/no question]

Example 2:  Yes?   [rising tone, answering the telephone/door]

The thing is, one-syllable utterances in English are quite rare.  So what happens when we string words together?  Well, we produce utterances, or “continuous piece[s] of speech beginning and ending with a clear pause” (Roach, 2013:315).

(It’s important to remember at this point that we’re talking about speech and sound. ‘Utterances’  are to do with phonology; ‘sentences’  are to do with grammar).

And when we look at utterances longer than one syllable, we realise that we can only really identify tones (pitch changes) on a few very prominent syllables.

So if we want to analyse the intonation across the whole utterance, we really need to talk about something bigger than a syllable.  For this purpose, linguists talk about tone units.

A tone unit isn’t simple to define, so let’s look at some illustrative examples instead.

Take the word ‘Japanese’.  Three syllables.  The last one is stressed; the first one is also slightly stressed.  So we can say that ‘-ese’ has primary word stress and ‘Jap-‘ has secondary word stress.  These are innate properties of the word, which we see marked in dictionaries with little lines at the top or bottom of the transcription, like this (from the Macmillan online Learner’s Dictionary):


If we simply cite this word (i.e. just say it, like a one-word utterance, as though answering the question “Is she English?”), it forms one complete tone-unit.  The syllable with primary word-stress is most prominent, and we can write it as follows:

// JApanESE //

(The convention in the literature is to represent tone units like this, with double slanted lines showing the boundaries of the utterance, single slanted lines showing the boundaries of tone units, CAPS showing stressed syllables and UNDERLINED CAPS showing prominent syllables.)

In connected speech, however, innately stressed syllables will not necessarily also be prominent.  This is because, unlike word-stress, prominence is not determined by the language itself but by the speaker (Brazil, 1994; Underhill, 1994).  It is possible for prominence not to occur on any of a word’s syllables, despite the stress(es) in its citation (dictionary) form.  For example:

// ACtually she’s japaNESE //

// a JApanese SHIP-owner’s been / KIDnapped //

// i thought SHE was japanese // NOT HIM

(McCarthy, 1991:95)

In these examples, we can see that word-stress and prominence may overlap, but innate word-stress is only automatically placed when a word is quoted as if from the dictionary, whereas in a tone unit, prominence depends on context.

Speakers naturally assign prominence as a matter of course, selecting what information to highlight as the most important part of their message and thereby focusing their listeners’ attention on a particular intended meaning (e.g. she in the last example above, as opposed to he).  Words which carry little information content, or which are predictable or  impossible/improbable to replace with anything different in the current discoursal context are therefore unlikely to be given prominence (e.g. the identity of she in the above utterance ‘Actually, she’s Japanese’ is likely implicitly understood by the speakers.

This brings us (finally!) to nuclear stress, also known as tonic stress.

The most prominent syllable in a tone unit – i.e. the syllable which carries the tonic/nuclear stress, also called the nucleus – is the point at which the major pitch movement happens, and the hearer’s attention is centred around it.

In other words, here’s what it all boils down to:

Placing prominence on a particular syllable draws a listener’s attention to it.  If a speaker places prominence on something he/she DID NOT INTEND to draw attention to, his/her listener is likely to be lost.

Isn’t nuclear stress a particularly native English feature?  Does it really matter in ELF interaction?

It’s true that a great deal of literature on English intonation and stress draws attention to native uses of these pronunciation features.  Some of these sources also imply the common assumption that native speakers of English will be the likely listeners to (and judges of) learners’ pronunciation; for example:

Pennington (1996:253) advocates teachers’ paying attention to stress and intonation as examples of

“general characteristics and overall voice quality which interfere with intelligibility or make the accent sound especially non-native, i.e. non-English or non-standard.”

Underhill (1994:75) also refers to the role of the ‘native listener’:

“What happens when a non-native speaker uses an inappropriate intonation pattern?  My observation is that while within certain limits of intelligibility mistakes or inappropriacies of pronunciation, grammar and even vocabulary can be accommodated by the native listener, inappropriate intonation can at times give rise not just to obscuration of the message, but to reception of a quite different message.”

But in fact, as we will see below, both native and non-native listeners alike can get a “quite different message” due to particular placement of nuclear stress; that is partly why this aspect of intonation features in the Lingua Franca Core.

Moreover, as this blog intends to reflect, many learners today are less likely to be interacting with native English speakers than with other people from different first-language backgrounds.  It is therefore no longer appropriate to assume a native-speaker interlocutor, or to take native-like pronunciation as the only/best goal or model.

Some material for teachers does recognise this state of affairs.  For example, Dalton and Seidlhofer (1994:iv-v) point out that:

“Knowledge about discourse tells us that appropriacy is a more important criterion for intelligibility than correctness.  This view accordingly shifts the emphasis away from native speakers as yardsticks of ‘correctness’ to teachers taking informed decisions as to what is desirable and feasible in order to meet the needs of specific learners.”

David Brazil also explains (1994:1-2), in his excellent coursebook for teaching intonation, that:

“[the book] does not follow the perfectionist tradition, which demands native-speaker-like control of the sounds of a particular accent, and which regards everything else as an ‘error’.  Instead, users are encouraged to see pronunciation from the point of view of how it can best enable them to make their meanings and intentions clear to a listener.”

Our concern on this blog is intelligibility in an ELF context, i.e. “any use of English among speakers of different first languages for whom English is the communicative medium of choice, and often the only option” (Seidlhofer, 2011:7); and Jennifer Jenkins included nuclear stress in her proposals for the Lingua Franca Core (2000), which is, of course, concerned only with intelligibility in ELF.  (Remember that criteria for a particular pronunciation feature’s inclusion in the LFC included whether it appeared likely to hinder intelligibility if used inaccurately and whether it was teachable.  Based on Jenkins’ data and teaching experience, she determined that nuclear stress met both these requirements.)

Let’s consider a brief example from her original data (Jenkins, 2000:50-51) which illustrates the importance of appropriate nuclear stress placement in ELF:

Four students, each from a different L1 background, were engaged in a typical classroom communicative task: creating a poster for the classroom wall.  One student asked the others // have you got a blue VUN//.  The others were lost.  This student was in fact referring to a pen, asking for ‘a blue one’.  When the speaker altered the nuclear stress to say //a BLUE vun//, the others understood; intelligibility was achieved.

In other words, this student had two pronunciation ‘errors’: misplaced nuclear stress and substituting /v/ for /w/ in the word “one”.  In the end, it wasn’t necessary to pronounce the word “one” differently to achieve intelligibility, but simply to place nuclear stress appropriately.  By making the word “blue” prominent, the listener understood the speaker’s intended meaning.

‘Nuclear’ stress, a.k.a…

Different linguists and literature use different terminology.  Sometimes the terms overlap; sometimes they don’t.  Let’s briefly examine some of the main contenders for potential confusion:

  1. Tonic stress – same as nuclear stress.  Confusion over!
  2. Sentence stress – more of a grammatical, rather than phonological, distinction.  The basic idea is that ‘content/lexical words’ (like nouns, adjectives and verbs) will usually be pronounced with greater stress than ‘function/grammatical words’ (like prepositions, articles and conjunctions).  For example, in the question “Do you want to go?” only “want” and “go” will be ‘fully’ pronounced.
  3. Contrastive stress – an example of how the effect of shifting nuclear/tonic stress can be exploited to change meaning by contrasting one word with another that could have been used in its place.  For example, “I love YOU (as opposed to some other guy/girl).
  4. Corrective stress – similar to contrastive stress, and typical of classroom dialogue!  Another example of how the effect of shifting nuclear/tonic stress onto a particular word can be exploited to identify a mistake in something that was said.  For example, “Oops, I meant NEXT month, not THIS month”.  (Aside: In the classroom, this sometimes has the amusing unintended inconsequence of a group of students repeating back to you something you’ve just corrected them on, oddly distorting the standard (citation/dictionary) form of a word.  For example, a student says “The film was really bored.”  You say, “borING!”  The student says, “yeah it was borING.”  You falter and eventually let it pass as they’ll forget the correct adjective ending in a minute anyway.)

How to teach nuclear stress?

See the second post in this series for some practical ideas.


Brazil, D. (1994). Pronunciation for advanced learners of English (Teacher’s book). Cambridge: Cambridge University Press.

Dalton, C. & B. Seidlhofer (1994). Pronunciation.  Oxford: OUP.

Jenkins, J. (2000). The phonology of English as an International Language. Oxford: Oxford University Press.

McCarthy, M. (1991). Discourse analysis for language teachers. Cambridge: CUP.

Pennington, M. C. (1996). Phonology in English Language Teaching. New York: Addison Wesley Longman.

Roach, P. (2013). English phonetics and phonology: A practical course. (enhanced EBook edition). Cambridge: Cambridge University Press.

Seidlhofer, B. (2011). Understanding English as a Lingua Franca. Oxford: Oxford University Press.

Underhill, A. (1994). Sound foundations. Oxford: Macmillan Heinemann English Language Teaching.



ELF in a multilingual class – finding common needs


As with all linguistic interaction, intelligibility is undoubtedly a two-way street – that is, it’s not only the speaker’s responsibility to make him/herself understood to others, but the listener also has the task of understanding.   In the classroom, we can work on both sides.  In other words, we can help develop learners’ productive and receptive competence.

In the multilingual classroom, however, this situation gets a bit complicated.

The problem

In my experience, if you have a classroom full of learners from different L1 (first language) backgrounds, they sometimes experience difficulties making their pronunciation intelligible to each other.

One problem lies with the listeners: they’re often unfamiliar with the pronunciation of English speakers from backgrounds other than their own.  Another problem lies with the speakers (and the teacher): how to address the particular difficulties of a student from an L1 background with which the teacher is unfamiliar?  Many English teachers I know speak another language, maybe even two other languages.  But it’s rare for them to be well-acquainted with ten different L1s – and yet this is just the situation they may face in the multilingual classroom.

The multilingual classroom is a ready-made ELF environment, but teachers may struggle to exploit this opportunity due to a lack of detailed background knowledge or suitable reference material that makes it reasonably quick and easy* to work out what to prioritise in a class full of learners from different L1 backgrounds.

The solution

This is the problem I set out to solve with my MA research, which I completed a few months ago.  I created a grid which featured details of the typical areas of pronunciation difficulty of 12 different L1 backgrounds, contrasted against the LFC.

The idea is simple:

  • find the L1s represented in your class on the grid
  • highlight those columns (or black out the others)
  • look across the selected columns to see where your students’ areas of likely difficulty overlap
  • this will quickly give you a shortlist of ELF pronunciation priorities to work on in class
  • you can decide how to rank the priorities according to how many of the students will have a particular point in common

The boxes on the grid are colour-coded (in greyscale) to show you how big a problem a particular area is likely to be.  For example:

Click to enlarge image

This means that aspirating this sound is generally difficult for Italian speakers.

Click to enlarge image

This means producing this sound is not usually problematic for Italian speakers (they just have trouble aspirating it, as is sometimes necessary in English).

Click to enlarge image

This means this sound is generally problematic for Spanish speakers, and ESPECIALLY (1) when it occurs between vowels and (2) the contrast with /v/.

Click to enlarge image

This means that producing this sound is not generally a problem for Spanish speakers, but that it can be difficult to pronounce accurately between vowels.


The 12 L1 backgrounds represented in the full grid are: Italian, French, Spanish, Russian, Japanese, Portuguese, Arabic, German, Polish, Chinese, Turkish and Hungarian.  (A list of primary sources on which the information in the grid was based is given below.)

An example in practice

Imagine I had 6 students in my class, whose first languages were French, Italian, Arabic, Portuguese, Chinese and Hungarian.

1. First I’d black out the other columns on the grid:

Click to enlarge image

(click to enlarge)

2. Then I’d look at the extent to which these learners are likely to have similar difficulties, and I’d work out my priorities.

When I do this, I tend to follow a ‘majority rules’ syllabus:

  • if the same area of likely difficulty occurs in 5-6 columns, we’ll definitely cover it in class
  • if something shows up in 3-4 columns, we’ll cover it in class but perhaps work on it less frequently, or more briefly, with an optional homework task for extra practice
  • if something shows up in only 1 or 2 columns, I’ll just keep an eye on the learners from those backgrounds during lessons and if it’s clearly a problem for them, I’ll aim to cover it in correction and speak to them individually about it so they can monitor their own speech.  (Of course, this is always the trouble with teaching groups – you just can’t cover absolutely every individual need in class time.  But you can prioritise according to common areas of need.)

My annotated grid might look like something like this (top priorities circled in red, medium priorities in orange, low priorities in blue):

Click to enlarge image

(click to enlarge)

3. Written out separately, this would be the resultant syllabus.

(I’ve included examples to illustrate features wherever I’ve used phonemic script which may not show up on some computers.)

High priorities (common to all, or almost all, students):

  • aspirated /p/
  • the velar nasal consonant /ŋ/ (as in ‘sing‘)
  • consonant clusters at the ends of words
  • long-short vowel contrasts (e.g. /iː/ and /ɪ/ in ‘sheep’ and ‘ship’)
  • /ɜː/ (as in ‘bird’)
  • shortening long vowels before unvoiced consonants (e.g. the vowel in ‘back’ is a bit shorter than the one in ‘bag’)
  • word grouping and placement of nuclear stress

Medium priorities (common to about half the class):

  • aspirated /t/ as in
  • aspirated /k/
  • /s/
  • /z/
  • /h/
  • /w/
  • /ʒ/ (as in ‘vision’)
  • /tʃ/ (as in ‘chin’)
  • /dʒ/ (as in ‘gin’)
  • consonants at the ends of words

Low priorities (specific to only a couple of students):

  • /p/ in general (specific to Arabic learner)
  • /b/ (Arabic & Chinese)
  • /t/ (French & Portuguese, and only in some contexts)
  • /d/ (Chinese & Portuguese, and for the latter only in some contexts)
  • /k/ (Arabic)
  • /g/ (Arabic & Chinese)
  • /f/ (some Chinese speakers)
  • /v/ (Arabic & Chinese)
  • /ʃ/ (as in ‘shoe’) (Chinese)
  • /m/ (French & Portuguese, and only in some contexts)
  • /n/ (Portuguese in some contexts & Chinese for some speakers)
  • /r/ (Portuguese & Chinese, for some speakers only)
  • /l/ (Chinese & Portuguese, and for the latter only in some contexts)
  • /j/ (Portuguese, and only in some contexts)
  • consonant clusters at the start of words (Chinese & Hungarian, and for the latter only in some contexts)

Great! Can I have a copy of that grid?

You can download the full grid here and the accompanying teacher’s notes here.  The files are in PDF format.  You can either view it on-screen or print it out – but note, it’s formatted for printing on A3 size paper (otherwise it’d be too small to read!).

Some more information about the LFC grid, if you’re interested

  • Academic literature on phonology and the Lingua Franca Core uses a lot of terminology which many teachers are not familiar with.  The terms in this LFC grid have therefore been standardised, simplified & defined to make it more accessible to teachers without extensive phonology training (e.g. ‘paragoge’ and ‘epenthesis’ have been avoided in favour of ‘adding short vowel sounds’; ‘tense’ and ‘lax’ vowels have been called ‘long/short vowels’, though I acknowledge the somewhat problematic/ imprecise nature of this substitution).
  • The information in the grid has been collated from many sources, some of which take ELF as their starting point and some of which don’t, some of which refer to particular varieties of an L1 (e.g. Brazilian Portuguese) and some of which refer to different varieties (e.g. European Portuguese).  Including the greyscale code allows the grid to be a bit flexible vis à vis what different authors have written.
  • Nuclear stress is an area which doesn’t receive a lot of attention in ELT training courses. We’ve covered this in more detail in another blog post.  For now, if you want to learn more, see Jenkins (2000:154-5) for more thorough discussion.
  • I originally made this grid as part of my MA research and piloted it at my own school.  So the 12 L1s included reflect the most common L1 backgrounds represented in classrooms at my school.  In future, I hope to extend the grid to include more L1s.

(*Earlier I said that any resource for teachers wishing to devise an LFC-based pronunciation syllabus needed to be reasonably quick and easy to use.  I’m aware of Learner English – but this is based on native-speaker norms, so a lot of the guidance is irrelevant for ELF purposes.  An excellent ELF-oriented resource for teaching pronunciation is Robin Walker‘s book, but it’s still a book, and takes some time to go through if you’re looking for what difficulties your students will have in common across several L1 backgrounds.  Sadly, most teachers I know don’t dedicate that much time to course planning, let alone pronunciation syllabus planning.  Hence, the grid format adopted here.)

Bibliography of sources used in compiling the grid

Baker, A. (1982). Introducing English pronunciation: A teacher’s guide to ‘Tree or three?’ and ‘Ship or sheep?’. Cambridge: Cambridge University Press.

Kallestinova, E. (2009). ‘Voice and aspiration of stops in Turkish’, Folia Linguistica, Vol. 38, Issue 1-2, pp. 117-144.

Kenworthy, J. (1987). Teaching English pronunciation. Harlow: Longman.

Rogerson-Revell, P. (2011). English phonology and pronunciation teaching. London: Continuum.

Siptár, P. & M. Törkenczy (2000). The phonology of Hungarian. Oxford: Oxford University Press.

Swan, M. & B. Smith (eds.) (2001). Learner English: A teacher’s guide to interference and other problems. (2nd edn.) Cambridge: Cambridge University Press. (VARIOUS CHAPTERS)

Varga, L. (1975). ‘A contrastive analysis of English and Hungarian sentence prosody’. Working papers of the Hungarian-English Contrastive Linguistics Project.  Published by the Linguistics Institute of the Hungarian Academy of Sciences and Center for Applied Linguistics.

Walker, R. (2001). ‘Pronunciation for international intelligibility’. English Teaching Professional 21.  Retrieved 22 Dec 2011 from

Walker, R. (2001). ‘Pronunciation priorities, the Lingua Franca Core, and monolingual groups’. Speak Out! The newsletter of the IATEFL Pronunciation Special Interest Group. 18: 4-9.

Walker, R. (2010). Teaching the pronunciation of English as a Lingua Franca. Oxford: Oxford University Press. (VARIOUS SECTIONS)

Zhang, F. & P. Yin (2009). ‘A study of pronunciation problems of English learners in China’. Asian Social Science, Vol. 5, No. 6, pp. 141-146.

Zoghbor, W. (2009). ‘The implications of the LFC for the Arab context’. Speak Out! The newsletter of the IATEFL Pronunciation Special Interest Group. 41: 25-29.


What is the Lingua Franca Core?


First, let’s recap:

The concept of ELF (English as a Lingua Franca) is simple: many learners of English today do not want/need to use English with people whose first language (L1) is English.  They are more likely to use English in situations where nobody shares an L1 (e.g. a native speaker of French, a native speaker of Japanese and a native speaker of Arabic might use English to communicate with each other).

Barbara Seidlhofer, a linguist at the University of Vienna who has studied and written extensively on this use of English, explains ELF as follows:

“any use of English among speakers of different first languages for whom English is the communicative medium of choice, and often the only option.”

From ‘Understanding English as a Lingua Franca’ , pg. 7 (reference below)

In this sense, people who speak English as their first language might also be present in ELF interaction, but they’d more than likely be in the minority.

What does this mean for pronunciation?

Way back in 2000, a linguist called Jennifer Jenkins wrote a book about this way of using English, including research on what aspects of pronunciation appear important to produce and understand accurately if a learner’s goal is NOT to sound precisely like a native speaker, but mainly to communicate intelligibly with other non-native speakers.

(More on the definition of “intelligible” in a future post… for now, let’s just understand it as the extent to which a listener understands a speaker’s pronunciation; and let’s just focus on the speaker’s side of things.)

One result of this research was the ‘Lingua Franca Core‘ (the ‘LFC‘).  This is a list of pronunciation features which appear to be crucial to produce accurately in order for ELF communication to be intelligible.  (There’s much more to it than this, of course – more detailed discussion will appear on this blog in due course!)

Outside the ‘core’ are all other features of pronunciation that might occur in different varieties of English; but these are probably not necessary for learners to be able to produce if their goal is mainly NOT to communicate with (or sound like) a native speaker.

By teaching features outside the ‘core’ receptively, rather than productively, learners can still understand other accents of English and maintain something of their L1 accent, which many learners may wish to do, given that accent is an important part of personal identity.

Jenkins’ suggestions for the LFC were very thoroughly explained in her 2000 book, which we’d urge you to read if you’re really interested in understanding the logic behind it.

In the meantime…

Here’s the LFC in short:

  1. Most consonant sounds + one vowel (/ɜː/)
  2. Preservation of most consonant clusters
  3. Vowel length (especially before voiced/unvoiced consonants)
  4. Appropriate word grouping and placement of nuclear stress

You may have noticed that some popular areas of focus in ELT (e.g. word stress, tone) aren’t included.  That’s not a mistake–those areas didn’t appear crucial to intelligibility in an ELF context in Jenkins’ data.  (More on this in later posts.)

And in more detail:

1. Consonant sounds

• All English consonant sounds are necessary EXCEPT /θ/ and /ð/ (for which most substitutions are possible, such as /f/ and /v/, but probably not /ʃ/, /ʤ/ or /z/).

• ‘Dark /l/’ (also written as [ɫ]) is not necessary.  Speakers can substitute ‘clear /l/’ (possibly preceded by a schwa if the /l/ is syllabic, like at the end of ‘bottle’).  Substituting /ʊ/ for /l/ at the ends of words might also be acceptable, but more research is needed to confirm this.

• /r/ should be pronounced as in General American pronunciation (technically called a “rhotic retroflex approximant” and written as [ɻ].  It should also be pronounced everywhere it occurs in spelling, as in American English.

• /t/ needs to be carefully pronounced between vowels (e.g. ‘Italy’) and in clusters in the middle of words (e.g. ‘winter’). It should not be ‘flapped’ (as in General American pronunciation, ‘Italy’ might sound like ‘Idaly’ or ‘latter’ might sound like ‘ladder’); and it should not be replaced with a glottal stop (like in Cockney ‘better’).

• The consonants /p/, /t/ and /k/ must be aspirated when occurring in initial position in a stressed syllable (e.g. the first /p/ in ‘paper’).

2. Consonant clusters

• Clusters of consonants at the beginning of words must not be simplified (e.g. learners mustn’t drop the /r/ at the start of ‘product’).

• Clusters of consonants in the middle or at the end of words are a bit more complicated.  They can be simplified if it makes articulation easier, but only according to rules of elision (i.e. dropping sounds) that also apply to native English varieties (especially in clusters containing /t/ and /d/, like ‘postman’).

• If learners have trouble producing consonant clusters, it’s usually OK to insert a very short schwa vowel between consonants, providing they don’t then stress this syllable (e.g. ‘product’ could be pronounced more like [pә’rɒdʌkәtә] by Japanese speakers without damaging intelligibility).

• Similarly, learners can add a short schwa at the end of a word ending with a consonant, provided this does not create another word which it might be confused with (e.g. ‘hard’ sounding like ‘harder’).

3. Vowels

• Length contrasts must be preserved, e.g. ‘pill’ versus ‘peel.  However, the actual quality of vowels is less important, providing it’s consistent (e.g. don’t keep switching between different pronunciations of the vowel in ‘hat’ so sometimes it sounds like RP [hæt] and sometimes it sounds like New Zealand [het]).

• The length of diphthongs must be preserved but, again, the actual quality of the vowels is less important, providing it’s consistent.

• When a vowel occurs before an unvoiced consonant, it should sound slightly shorter than when it occurs before a voiced consonant.  For example, the vowel in ‘right’ is slightly shorter than the vowel in ‘ride’, and the vowel in ‘kit’ is slightly shorter than the vowel in ‘kid’.

• The /ɜː/ vowel, as in ‘girl’ or ‘first’, must be pronounced accurately.

4. Word groups and nuclear stress

• The stream of speech should be divided into meaningful tone units (also known as ‘tone groups’, ‘word groups’ or ‘thought groups’).

• Nuclear stress (i.e. which word is stressed within a ‘tone group’) must placed appropriately, especially for contrast/emphasis.  This means the difference in meaning should be clear between, for example, ‘Let’s meet NEXT Saturday’ and ‘Let’s meet next SATURDAY’.

How to practise features in the LFC?

In due course, we’ll be posting practical resources on this blog for working on these areas with learners whose main aim is to use English as a lingua franca.

One such resource, focusing on just one part of the LFC (consonant sounds) is already available: Battleships.

Further reading

Jenkins, J. (2000). The phonology of English as an International Language. Oxford: Oxford University Press.

Jenkins, J. (2007). English as a Lingua Franca: Attitude and Identity. Oxford: Oxford University Press.

Seidlhofer, B. (2011). Understanding English as a Lingua Franca. Oxford: Oxford University Press.

Walker, R. (2010). Teaching the pronunciation of English as a Lingua Franca. Oxford: Oxford University Press.