Cracking the code

Alan Turing would have celebrated his 100th birthday this year.  Turing is mostly remembered for his work on cracking the Enigma code. For some learners authentic material in English is just that:  an enigma. At best difficult to understand, at worst an impenetrable mystery.  After hearing Mark Hancock speak at TESOL France’s Annual Conference I felt inspired to try some of Mark’s recommendations to help learners decode authentic material.

Many people believed Apple’s logo to be a tribute to Alan Turing

Mark Hancock’s presentation, which you can access here, looked at the reasons why some learners find authentic speech difficult to follow, and what can be done to facilitate listening with more ease. In his presentation Mark talked about two main roadblocks for listeners. Firstly, the problem of connected speech (where does one word stop and the next start?) and secondly the  sheer variety of accents, and the resulting variation in pronunciation. For some learners, there is a third problem, and that is within the word itself. Some learners are not able to distinguish between certain sounds, making it impossible to know if a speaker is talking about ‘some flowers‘ or ‘sunflowers’, for example.

I designed the lesson to focus on these three areas. I was interested in helping learners to raise their  awareness of these barriers. I also wanted to provide some helpful and comprehensible input  that would add to their corpus of language. My aim was to facilitate the development of ‘mental pathways’, routes for ‘information’ to follow as it journeys from ear to brain, where it can connect with an appropriate meaning.


The learners were a group of seven adult women attending a one day course aimed at improving listening skills. They all worked at the same company but none of them knew each other. They were intermediate learners, whose exposure to English ranged from working in an international team with English as the Lingua Franca, to using English once or twice a week on the telephone. They all had French as L1 and none of them were bilingual in another language.

I won’t give you a blow by blow account of the lesson, but some of the things I observed may be of interest to you:


I played the Beatles #1 album as background music and handed out sheets of A4 paper each with a different question on them. The learners read the question, wrote their answer and then passed the question to the next person. This enabled the learners to express their objectives, list the types of accents they found difficult and give examples of listening they were exposed to each week (both for pleasure and work). To do this they did not have to decode any sound, the inputs and outputs were silent.

I had been playing ‘Yesterday’ during the opener. I then followed Mark Hancock’s lesson plan for his ‘Wrong Lyrics’ lesson.

Here are my observations from this part of the lesson:

The activity allowed the learners to notice some of the challenges to listening. They identified linking, elision and assimilation.

By correcting the print out of the wrong lyrics they could see and repair the language.

The visuals from the video provided a powerful hook to link the problem and the effect, for example later in the lesson the learners would comment that there was a ‘chateau’ problem in the audio.

It was very entertaining, learners realised that anyone can mishear and it lightened the  weight of the ‘ listening problem’

Using their observations we went to work on other sentences with examples of elision, linking and assimilation. The learners were able to identify what was happening to the words as they were used in connected speech. We worked on a whiteboard and annotated the chunks to show what was happening to the sounds. Although the focus of the lesson was ‘listening’ all of the learners automatically produced the speech as part of the learning process. In Mark’s words:

“there is no better way for a learner to become aware of something receptively than attempting to actually produce it”.


There’s a chateau hanging over me

Early in the lesson I had suggested that we do a listening ‘test’ to find out if any sounds were particularly hard for them to differentiate. All of the learners agreed to this. For this I used the first test from Ann Baker’s book, Ship or Sheep.

Although there was a wide range of responses certain sounds emerged as common problem areas:

/n/  – all of the learners – ‘some flowers’ – ‘sun flowers’

/ʤ/ – five of the learners – ‘jeep’ – ‘cheap’

/ɑ:/ – five of the learners –  ‘hat’ – ‘heart’

/h/ – five of the learners – ‘ill’ – ‘hill’

We worked on these phonemes at various points during the lesson, listening to minimal pairs and using ‘acoustic drilling’ style exercises. I deliberately spread these throughout the day, so as not to overload the learners. The learners were enthusiastic about these exercises and expressed relief at being able to correctly identify the sounds. The learners were able to see their own results and several learners were interested in self-study in this area after the course.

If anyone has any insights or observations around these clusters of problem sounds in terms of sound reception I’d be very glad to hear about it! They seem to be ‘typical’ problem sounds for French speakers but I’d love to know more if anyone has any ideas.


Ship or Sheep, Ann Baker, Cambridge 

The afternoon part of the lesson was focussed on listening materials using accents they had identified  as difficult to understand in the morning session. The most frequently encountered accents were Indian and South African.

I had expected Indian to appear, and had previously prepared some micro-listening based on the Collins English for Business Listening book by Ian Badger. I used Audacity to make very small and repetitive clips of ‘vulnerable’ sounds in action.  Some of the words I had chosen were:

Corporate – vulnerable /r/ is dropped, sounds like ‘coprit’

Rural  – middle /r/ has a ‘w’ quality

Think – /θ/ changed to a ‘t’ – sounds like ‘tink’, even ‘dink’ depending on speaker

Difference – /ɪ/ changes to /i:/ – dee -fer -ents

I used three of the audio files from the Collins listening book, and one video from on the rise of cricket


The learners were able to ‘pick out’ the words that were different to their expectations on the first or second listen.

The learners were able to identify what was happening to the words and transcribe this using their observations from the earlier part of the lesson.

After the micro-listening, the learners were not only able to recognise similar patterns in subsequent audio inputs, they were also able to mimic the sounds. (This occurred spontaneously, and seemed key for the learners to try to make the sound in order to connect with it).

The learners were then able to recall other words which they thought had sounded different to their expectations. One was ‘solution’ which they described as having an extra ‘y’ ‘sol -iyu-shun’.

To cover off the South African accent I used another listening from Collins. I had not anticipated this accent, but I was able to generate a sort of micro listening by repeating the same stretch of sound over and over. The learners picked out the changes in vowels and noticed the similarities in what was happening with /θ/ and /r/ between the Indian and South African speakers. (We repeated a chunk which mentioned the time ‘four thirty’)


Throughout the lesson I had asked the learners how they felt about the audio. I asked them to rate the difficult on a scale of 0 – 10, where 0 was no problem and 10 was extremely difficult. As the lesson progressed the first listen descended from being around 8 for most learners, to being around 4 or less by the end of the lesson.

The learners were also asked  to fill in a feedback form and to write down three things they had learned during the day and two things they thought they would do differently when listening in the future.

All of the learners reported that they felt ‘more at ease’ or ‘much more at ease’ with the subject

6 of the learners gave the lesson 5/5 for quality of teaching and 5/5 usefulness of the materials and content used. (The other learner rated the lesson 4 and 4, commenting that she would have like to have covered the Chinese accent)

The comments that came back from the learnings were around listening to the key words, understanding better what was ‘happening’ to words in connected speech and realising that accents were not as big a problem as they thought.

When I reflected on the lesson afterwards, I was quite happy in general with the outcome for the learners. Obviously there were things I thought I could have improved on, but I was happy to have taken the risk and tried something new.

One of the things that bothered me was about ‘systems’ and ‘skills’. I remember being taught that listening is a ‘skills’ lesson. But this to me felt much more like a ‘systems’ lesson. In fact it felt like teaching pronunciation in reverse. I wasn’t hoping for an improvement in production or sounds but in ‘reception’ of sounds. When I looked back on the lesson, it had more of a ‘systems’ structure to it than anything I had taught before in a skills lesson.

I would be delighted to hear your observations or experiences in teaching in this area. Please leave me a comment if you have a moment to spare. If you are interested in listening and this type of approach there is a fascinating article by Sheila Thorn on this subject.

There were lots of comments from the learnings feedback. One learner commented that she felt that Indian accents had become less of a mystery, and she felt she could now follow speakers more easily. I was pleased that she had cracked part of the code, and thought back to Alan Turing. Although he was brought up in England, his parents met in India, and Turing’s ancestors can be found in India as early as 1700. Fitting somehow that this was the accent the learners decoded in the lesson.


4 thoughts on “Cracking the code

  1. hi carolyn
    really liked the post (especially way u used codebreaking and turing) and have been doing accent work with french learners myself some weeks back (

    are you aware of the book “learner english” edited by swan and smith? there’s a good chapter on diff btw french and english on vowels, consonants, consonant clusters, stress, intonation, and influence of spelling on pronunciation.

    but it has no info on connected speech.

    in your micro-listening part how many words in total did you use? do you think allowing the learners to be able to listen individually and control the playback would increase the effectiveness of this technique?


    • Hi Mura,

      Thanks for your comment and book tip! I’ve had a look at ‘Learner English’ and it does highlight some of the problems the learners had as being typical. It’s an interesting book – great tip.

      We didn’t do too many micro listenings. I had prepared the following in advance:

      Dropped T – ‘most extreme cold’ ‘most extreme warm”don’t normally’
      TH – ‘nine thirty’ ‘I think’
      R – ‘corporate’ ‘rural’
      I – ‘different sides”different terrains’

      The ones from the TED talk we looked at by going back over the video. We did the same for the South African accent. The learners were able to draw parallels between the two accents, the ‘th’ and the ‘r’ especially.

      I think it would have been great if they could have controlled the play back individually, but it did work very well as a group activity too. The learners seemed reassured by the fact that they were not alone, and that by hearing the ‘chunk’ in isolation several times they were able to focus on identifying the ‘changes’ that were happening to the expected pronunciation when the accent is applied.

      I was taken aback by just how effective these two activities were. I was ‘test driving’ a technique from a conference, and I was really pleased with how much the learners got from it.

      If you give it a try please do let me know how it goes!


      • thanks for the info, very useful.

        accent work usually crops up as a class goes so for me preparing a microlistening may not be worth the effort. and simply rewinding to a problematic item may be more efficient. but certainly for a class dedicated to accents microlistening would be worth it.

        i guess using the learner english book and our classroom experience building up a database of microlistening may be an idea. though again how efficient this would be is debateable e.g. use of words outside of any context?

        also some research shows that grammar and vocab may be more important than accent

        thanks again for a great post

  2. Pingback: Using the IDEA accent archive in one-to-one classes « EFL Notes

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s