On 9 September 2017, Laura Patsko presented the following IATEFL webinar:
Teaching pronunciation for English as a Lingua Franca (ELF)
What do learners of English need to sound like? Who do they speak to? Who needs to understand them? Who do they need to understand? In 2017, the answer to all these questions is probably not “native English speakers”. Linguists estimate that non-native speakers of English now outnumber its native speakers by at least three to one (Crystal, 2008), and approximately 80% of interaction in English worldwide takes place with no native speakers present (Beneke, 1991). What does this mean for our classrooms? This webinar will consider new pronunciation priorities and challenges for learners and teachers of English, including practical tips and activities.
If you would like to watch the recording, you need to be a member of IATEFL (find out how to join), but even if you’re not a member, you can read Laura’s answers to some of the questions below. Over to Laura…
Thanks to all those who were able to attend my webinar on 9 Sept 2017. There were so many great questions and not enough time to respond to them all! So in this post, I’ll address some more. People often asked questions on the same general themes, so my responses are grouped accordingly. I’ve also included links to further resources so you can continue exploring ELF and pronunciation.
First of all…
Bachir Sahed: Do we teach our students to be EFL or ELF speakers?
It depends on their needs. One argument is that teachers should determine whether their students are really learning English in order to use it as a ‘foreigner’, i.e. living in a country where most other people use English as their first language, or to use it as a lingua franca, i.e. interacting in English with people from many different first-language backgrounds. They should then teach the students appropriately according to the different priorities of these two different use cases. Another argument is to recognise that we rarely fully understand the future needs of our students (and they themselves might not know!), in which case we should trust in statistics and assume that most people learning English today will need ELF, so that should be our focus.
My own feeling is somewhere in between. I think the most effective users of any language are those who appreciate that language is inherently diverse, flexible and changeable, and who have enough awareness of themselves and others to modify and moderate their use of language to suit different contexts and audiences. In practice, this means that our students today will need a sophisticated sense of what is likely to be most widely understood by the people who they are likely to interact with, but they will also need to be prepared to be surprised and to adapt to new contexts and interlocutors, which they will certainly encounter one day.
Apapan Ruengkul: Do you think that teaching pronunciation for ELF and EIL is the same?
Generally, yes. As research continues and our understanding of language phenomena develops, different scholars propose different terminology. What was once called ‘EIL’ [English as an International Language] is now generally called ‘ELF’ [English as a Lingua Franca], but even the term ‘ELF’ is now being discussed and sometimes replaced by other terms in more recent research. You can learn more about these developments in a free online course developed by the British Council and the University of Southampton’s Centre for Global Englishes.
This is closely related to…
What learners need/want
Arnab Podder (India): In my context (India), I have many students asking me “How do I master X accent”? What should be an appropriate way to deal with it?
We need to understand our students’ reasons for wanting to acquire a certain accent. Do they understand the implications of different accents? Do they (mistakenly) equate a specific accent with perfect international intelligibility? Do they actually want to speak with a certain accent, or do they really want the prestige associated with that accent? Do they simply like how a particular accent sounds because they’ve heard it in films? And so on. We need to have an ongoing discussion with learners about why they want what they want. To my mind, this is what distinguishes English teaching from English education. Of course, once our learners are able to make a considered, informed choice, then it’s our job to help them – by providing guidance, support and resources. Just as we would do for any other aspect of their English learning!
Some of you also asked about…
The content/focus of pronunciation instruction (especially for ELF)
Paul Seligson: Given the large number of cognates between English and Romance languages, where placing the stress in the ‘right’ place will effectively ‘gift’ you an English word, why is word stress not considered important in ELF?
It’s not that word stress isn’t considered important, just that from existing research word stress still appears to be a ‘grey area’ which needs further investigation. My own view is that knowledge of word stress will be useful for producing and understanding nuclear stress*, which is widely agreed as important for intelligibility. But this is a different justification for teaching word stress than the one you suggest. There may well be many cognates between English and Romance languages, but of course Romance languages are not relevant to all ELF research, or indeed all English users.
One thing that most pronunciation researchers agree on, and particularly those focusing on ELF, is that a learner’s L1 is one of the most helpful indicators for identifying priorities for pronunciation work and guiding the learner to producing particular pronunciation features. So if you’re teaching someone whose L1 is a Romance language and you are aware of cognates with only slightly different pronunciation, I’d certainly agree that it would be useful to draw their attention to this. But I couldn’t realistically give the same advice to all teachers and learners of all different L1s.
*Claudia Camenzind: What is nuclear stress?
I’ve written an introduction, with links to classroom activities, to this topic here.
Gonzalo Eduardo Espinosa: Would you suggest working on suprasegmentals first and then segments? Or both at the same time?
This is a difficult question to answer. Different learners have different needs and priorities, so I would probably give different advice to different people! My own preference is usually to start with segments and build them up into increasingly larger structures (syllables, then polysyllabic words, then phrases, etc.). But even more important than this is to simply include pronunciation with any new language structures that we teach. So if we’re teaching vocabulary in one lesson, we should also teach how the words are pronounced (i.e. sounds and syllables). If we’re teaching functional phrases in another lesson, we should also teach where the nuclear stress would likely fall in these phrases (i.e. a suprasegmental feature). And so on.
An interesting topic which I didn’t cover in the webinar was…
The role of ELF and pronunciation in assessment
JJ Polk: How does intelligibility of ELF play into standardized exams today?
James Easton: What steps are being taken to question the way English pronunciation is currently assessed by oral examiners for Trinity College London, Cambridge and IELTS etc?
To the best of my knowledge, the way pronunciation is assessed by standardised test examiners isn’t an immediate concern to the test providers. Pronunciation is perceived by so many in the ELT industry as esoteric and low-priority. Moreover, assessment of speaking in standardised tests tends to have a three-way tension between the need for consistent benchmarking, the need to be seen as respecting the diversity of English(es) worldwide and the need for assessors to be able to apply their discretion when exercising their specialist expertise and assessing someone’s oral intelligibility. These factors together give assessors a lot of individual power but little guidance in how to exercise it.
For example, the need for consistent benchmarking tends to lead to a (perceived or assumed) need to specify a standard norm to mark against, yet no exam boards seem to want to make an explicit choice of which accent assessors should use as a norm (for example, Received Pronunciation), presumably to avoid unpopularity among all the millions of speakers who don’t themselves know or use that variety. This leads to many exam descriptors that are unhelpfully vague or presumptuous, for example:
[pronunciation] is easy to understand throughout; L1 accent has minimal effect on intelligibility
– publicly available descriptor for IELTS band 8
Easy to understand by whom? The examiner? An imagined interlocutor? In what context? Why assume that an L1 accent will have an effect on intelligibility?
[candidate can] Produce individual sounds so as to be fully understood by the examiner, with only a rare sound that deviates from an internationally intelligible model
– publicly available descriptor for Trinity GESE band 12
At least here, a specific listener-judge is identified. But what counts as an ‘internationally intelligible model’? There is no widely agreed such thing. Such descriptors attempt to show awareness of international English without actually providing anything of practical use. It’s an exercise in ‘box ticking’. The best available evidence-based set of pronunciation features that enhance internationally intelligibility is arguably the Lingua Franca Core (LFC), but this is not a model per se. And in my experience, it’s unlikely that these examiners are trained to know the contents of the LFC. What’s more, a listener’s expectations and experience have a significant bearing on whether a speaker’s pronunciation is intelligible. Yet in an assessment situation, only the speaker (the person being assessed) will suffer the consequences of this mismatch in listener expectations and speaker performance.
Ultimately, the best judge of whether someone’s pronunciation is internationally intelligible is someone who is interacting with that person in an ‘international’ context. In other words, an individual examiner with little knowledge of ELF research and implications, who is forced to make a judgment in a few minutes based on an interaction between the test candidate and only one other speaker (listener), is probably not the best assessor – but it’s the best solution we’ve got at present given the logistical practicalities of assessing thousands of test takers worldwide year-round. Certainly, human markers are still much better than machines for comprehending speech and assessing speaking skills. Going forward, I would advise that test descriptors are revised to focus on intelligibility over acceptability, and then ensure that assessors are better trained in how this is actually likely to sound in practice.
There was a programme on BBC Radio 4 several years ago in which they touched on this subject, if you’d like to hear what someone from an examining body has to say.
Now, coming back to the classroom, some people had questions about…
Minimal pairs vs. context
Kemal Bereksi: the minimal pair ‘had’ : ‘hat’ could be confusing in isolation, but will hardly ever be confused in context. Don’t you think so?
Dong An: Would minimal pairs only be useful if the two words are of the same type? They wouldn’t confuse an adjective with a verb, would they?
There are really two issues here: one is about the way words in context might (or might not) confuse us, and the other is about the purpose of minimal pairs.
First of all, pronunciation does not operate in isolation. If a speaker deviates from norms and/or listener expectations in lexis or grammar, this can be compounded by pronunciation factors. So yes, in theory, many minimal pairs would be easily distinguished by the lexico-grammatical context. But in practice, when we listen to someone speak, what we’re actually doing is very rapidly decoding the sounds they produce, matching these strings of sounds to words in our mental lexicon, then matching these strings or words to grammatical patterns, so that we can understand the whole discourse. If that first step – decoding – doesn’t happen easily, each subsequent step can be slowed or stopped. Meanwhile, the speaker continues speaking and the listener is left behind. It’s no longer a question of identifying whether a word is an adjective or a verb, because there was already an issue in the first place with recognising the sounds. Certainly, listeners do use contextual clues to understand what they hear, and this can often resolve potential breakdowns in understanding. But even proficient listeners and speakers can occasionally misunderstand each other, so it’s helpful for learners to raise their awareness of meaningful contrasts through minimal pair work in class, so that they are better equipped to self-monitor during future interactions.
And secondly, even this explanation still neglects the fact that minimal pairs are probably more useful as a teaching technique for pronunciation than as an example of confusion risks. Using minimal pairs in class is just a way of illustrating contrasts that otherwise would be completely abstract for learners. Thus, the ‘hat’/‘had’ example is just an illustration of one particular phonetic contrast that is important for international intelligibility.
And finally, a number of you asked if I could recommend…
Christina Cacha: Can you recommend a good book that provides pronunciation activities?
I feel obliged to promote my own co-blog, which is full of activities for pronunciation and listening for ELF! The website is http://elfpron.wordpress.com. After that, my favourite activity book is Mark Hancock’s Pronunciation Games (CUP, 1995) and my favourite book about pronunciation and ELF specifically is Robin Walker’s Teaching the Pronunciation of English as a Lingua Franca (OUP, 2010).
Jani Pandiri: Is there any material available where speakers of different countries use the same text?
Yes, there is a website called the Speech Accent Archive: http://accent.gmu.edu
Thanks again for contributing to the ever-fascinating discussion of pronunciation for ELF.
I hope to see you at the next webinar!
Formerly an English teacher and teacher trainer, Laura Patsko now works as Senior ELT Research Manager for Cambridge University Press. She holds a BA in Linguistics and an MA in ELT & Applied Linguistics, and is particularly interested in the use of English as an international lingua franca, teaching pronunciation and investigating the practical applications of linguistic research. She blogs at laurapatsko.wordpress.com and elfpron.wordpress.com, and tweets as @lauraahaha.
Thank you to Laura for agreeing to write for the IATEFL blog.
If you’d like to write a blog post or present a webinar for us, please contact blog (at) iatefl (dot) org.
You can find out more about upcoming webinars on the IATEFL website. If you are an IATEFL member, you can access the recordings and slides from all of our webinars in the members’ area. If you’re not, you can join here.