Since I did an assessed lesson based on listening for my Delta, I’ve been interested in finding out more ways to help my students develop their listening skills. I even did a presentation on it myself at IATEFL Harrogate 2014, heavily influenced by John Field’s Listening in the Language Classroom [affiliate link]. Richard Cauldwell’s book Phonology for listening: Teaching the stream of speech [another affiliate link!] has also been very useful in helping me to understand why it can be so difficult for learners to decode fluent connected speech. I was therefore very pleased to be able to attend presentations given by both John and Richard during this IATEFL. The other presentations summarised in this post were from the Forum on Listening.
Still my favourite listening picture…
I haven’t tried to summarise Jane Setter’s plenary on intonation for two reasons: 1. I couldn’t get it all into tweets, 2. to fully appreciate the intonation differences, you really need to watch/listen to it yourself!
You can also watch an interview with Jane recorded after the plenary.
I definitely feel like I understand intonation and how to teach it much better now! Sue Swift also wrote about the plenary at ELT Notebook.
Listening: ways out of the fog (John Field)
John started by telling us that he has been ‘worrying about listening for the last 34 years’.
The typical comprehension approach starts with pre-listening and activating schemata, something that doesn’t reflect real life. We also don’t pre-teach vocabulary in real life, so it’s counter-productive in training. The next step is to set questions to target listening, then play the recording, check answers, and replay key sections (providing answers as we go). However, this approach doesn’t actually train learners to become better listeners.
(not a great picture, I know, but it shows why the student thought the answer to a question was ‘crack’, when in fact it was ‘nature reserves’ – they misheard the word ‘attract’ in the stream of speech)
John thinks that we can get a lot more out of the comprehension approach. We should forget about activating schemata and pre-teaching vocabulary, as these don’t happen in real life, though we should quickly introduce students to the context and the number of speakers, as they would normally know this, e.g. two people talking on a bus.
It isn’t usually the script that causes the problem, it’s the recording. The item writer is not involved in the recording. Tasks also need to be carefully thought out, as we give away in writing a lot of what’s in the recording, and encourage test-wise strategies, rather than strategies for general listening. To improve their approach, the teacher needs to prepare the listening in detail – using the script WITH the recording. What’s perceptually difficult?
There should be a first play for SS adjust to speakers’ voices, listening globally without the pressure of questions, and without using questions to guess in advance what they’re likely to hear. If you’re going to use questions, set them before the second listen, so that they don’t interfere with the learners’ perceptions the first time they hear the text. Check and DIAGNOSE reasons for learners’ answers, then replay parts identified by you AND students as perceptually difficult. Transcribe short sections, especially if they are particularly problematic. Listeners are individuals, each with their own problems with a recording, and these change over time and with experience. If you can afford it, provide a listening centre where listeners can work on their own problems individually. Give learners a transcript at the end of the lesson and ask them to listen again. Set listening homework. Though my students almost never do it – I need to have more engaging tasks and clearer developmental reasons for them to do it!
Do we need better materials? YES! New writers often don’t have a clear idea of what a skills approach is. Materials should take pressure off the teacher, guiding them, and focussing on the difficult parts of the recording. Field suggests open-ended questions, so the class can talk about possible answers. Sometimes you could use short clips with a single question, instead of the more normal long excerpts with multiple questions. You could also embed oral questions in the script rather than written questions to stay within the same modality (so have the speaker ask a question, then pause for answer). Materials should demonstrate a better understanding of the processes that underlie successful listening and design questions to target them.
We also need to understand better what it is we’re actually teaching! When you listen you:
- decode sounds
- search for words
- parse (grammar)
- construct meaning
- construct discourse
Sometimes context makes us distort the phonemes we think we’ve heard to make it fit the context. To handle the speech signal, we have to adjust to speaker’s voice (pitch, speech rate etc), then match the set of squeaks/buzzes we hear to the sound system of English. But do phonemes really exist? They are so variable, maybe they don’t. This echoes Richard Cauldwell – see below. To handle words, we have to divide connected speech, recognise spoken word forms, link them to what’s know about the topic. To parse, we must hold in our heads the words which have already been said, recognise grammatical patterns, and work out the word’s sense in it’s co-text. To construct meaning, we have to put what we’ve heard into a wider context, interpret new information in relation to this, infer information the speaker has taken for granted, and link words like he/she/it to what they refer to. This is followed by putting it into wider discourse (I couldn’t keep up at this stage!) There’s an awful lot going on!
How does this knowledge help us? An expert listener does these things automatically, but L2 listeners need lots of effort to do each of these things, so it can be hard for them to form the ‘big picture’ of what they’re listening to. Up to about B1, learners have to give a lot of attention to decoding at word level, limiting their ability to tap into wider meaning (I’ve definitely found this with Polish). Strategy instruction should therefore mainly be done with lower-level learners to equip them with the fact that they often can’t make wider sense of what they’re hearing. Strategy training helps them with real-world situations and to compensate for gaps in text. Lower-level learners needs process training and strategies training to fill in gaps.
You can use the same audio, but vary the task to target any of the five levels of listening.
We should also vary the levels of our strategies instruction:
A syllabus for listening: less top-down! More bottom-up (Richard Cauldwell)
Richard relishes fast, messy speech and tries to find ways to help learners understand it. His CoolSpeech app was an ELTons 2013 winner for digital innovation, and he is currently in the process of writing a follow-up to Phonology for listening: Teaching the stream of speech [affiliate link] which will provide a clear syllabus for listening for language teachers to work from.
All words have multiple sound shapes. Decoding is the skill of recognising words in the sound substance (or ‘fog’ in John Field’s words). The sound substance is the acoustic blur of speech, which exits the mouth, travels through the air, and hits the ear. It’s what exists before perception.
Richard uses a metaphor for different speeds and qualities of speech. The greenhouse is the place for citation forms. In the garden, sounds touch each other gently through the basic rules of connected speech. In the jungle, wild things happen and all bets are off. He argues there are different goals for pronunciation (clarity/intelligibility) and listening (understanding fast, messy, authentic speech).
Teachers tend to brush the mess of sound under the carpet, so even CPE Grade A students have ‘can’t do’ listening points. Every word has a ‘word cloud’: a range of possible word shapes in fluent speech, of which the citation form is the least likely.
The examples of ‘and’ above were all taken from a single conversation, with only one instance of it even vaguely approaching the citation form.
Here are just some of the changes which can happen in fluent speech (in the jungle):
- Consonant death: this can appear in many ways e.g. that changes to ‘at
- d’eth drop: anenatwasat– no ‘th’s (instead of and that was that)
- B-drop/B-soft – often happens with adverbs e.g. superbly – the ‘b’ can be lost or very soft
- Smoothie: when diphthongs/long vowels change to just one of their elements: like > læ
There were a lot more, but I just couldn’t keep up!
To help student, we need to delve below the word that is meant to the sound substance itself – what sounds were actually produced. Field and Thorn both advocate using short clips to help students focus (see ‘the bathtub experiment’ below too). Audacity is the best tool to help you break up the stream of speech.
Try this activity: replace the ‘i’ with each of the vowel sounds in the image. This helps students to prepare for different possible ‘shapes’ and accents:
Or take a phrase for a walk in the jungle, and give your students an earworm to take home with them:
The earworm should be short term, memory length, annoying, and stick in their head to prepare them for perception. I sometimes wake up with words or phrases like this in my head from foreign languages.
Have students listen along with an audio, and when they get to the most important part of it (the ‘wave’), get them to speak along with it (‘ride the wave’):
If you’d like to find out more, have a look at Richard’s website.
After the conference, I noticed that Richard shared Tubequizard as a link on his handout – it’s an excellent way to help students to focus on connected speech.
Adventures in listening: the bathtub experiment (Marie Willoughby)
Marie teaches students who attend full-time classes at IH London. She finds her students only get so far with listening before they begin to disengage. Sheila Thorn inspired Marie to start trying different ways of approaching listening in the classroom, and after listening to an episode of the The Moth podcast, she realised she had the perfect material to use, talking about a man who sailed across the English Channel in a bathtub. The only problem was that it was 17 minutes long! Marie decided to break it into a series of mini episodes, each with a cliffhanger. She then used these over a series of lessons.
Marie found that the best texts to use as episodes have a clear narrative, but are outside normal experience. In this case, pure sound is better than video, as students are more invested in understanding what they’re hearing, and will therefore try harder to apply the decoding strategies. If they have pictures to help them, they don’t need to work so hard to apply the strategies. Once she had the episode, she asked two questions: What stops them underrstanding? What will help them understand later?
First, they always listened with no task, then worked together to co-construct the text with other students. They then moved on to focussing on a particular decoding strategy, which students were then able to apply in later lessons. By using short excerpts in small chunks you have time to pause and get students to consider the language in more detail. For example:
- Decode past perfect v. past simple when listening, first as a gapfill with a section they were familiar with, then listening to the next part of the story and saying what they heard.
- Say why a speaker would choose one particular phrase, or why they would repeat it.
- Listen to his description of a problem. Draw it on mini whiteboards.
- The storyteller talked about English/French attitudes to each other, so Marie asked them to research it, after which the students understood the jokes better.
- Vocab was a problem, especially familiar words in a new context, so she got students to listen and complete a sentence, then think about the meaning.
What are the benefits of using an ‘adventure in episodes’? There’s no need to reset context or activate genre knowledge each time. Prediction is a natural part of listening to such a story. The students were really motivated to find out what happened next. Intensive decoding work really bore fruit – they were invested in doing detailed intensive listening work. It also developed their autonomy.
I really like this idea, especially for summer school or 121 lessons, though I think it could take quite a lot of work to prepare. It reminds me of my French teacher introducing us to the French musical version of Roméo et Juliette, with us listening to one song each week and trying to follow the story. You could also use the BBC Short Cuts or Listening Project podcasts. Of course, once you have it for one adventure, they don’t date much, so you should have it for the future. If teachers share this kind of thing on blogs, you could have examples to draw from. If you choose to do this, please share the link below!
Listening using smart devices: effects on student interaction and autonomy (Clive Shaw)
The conventional classroom layout has speakers at the front, but Clive wanted to know what happens if we change where students/teachers are in the room. If the teacher is in control, it’s not easy to monitor, and students don’t have much of an opportunity to work on their own listening strategies. It’s also difficult for students to transition from the controlled environment of the classroom to the unguided environment of the lecture hall (Clive works in EAP – I think this is true all the time, and this prompted by IATEFL Harrogate presentation).
Clive investigated how listening from a smartphone changed the dynamic, encouraged autonomy and gave students the opportunity to employ strategies. He designed the materials based on two sources: taken from YouTube or by creating his own recordings. To get the recordings to the students he used TubeChop for YouTube and Audioboom for his own recordings, then a link shortener (my preferred one is bit.ly). Recognising that these took longer to prepare than coursebook audio, Clive deliberately selected easy-to-prepare tasks, for example two-column notetaking.
The biggest difference in the classroom was the seating plan:
Each group had one phone, and the weakest student was normally in control of it. Clive found out that students became more aware of listening strategies they employed when using smartphones. Students were more able to use context/syntax to decode problem areas as they could play it again as many times as they wanted to. Students were also encouraged to make their own decisions about when and how to review extract.
I’ve always played with seating arrangements, but it had never occurred to me to do it with listening extracts before. This seems like a great way of helping mixed-ability groups in particular, and also helps students to get used to background noise when listening, something we don’t do enough of.
Tweets from other sessions
Here’s Pete Clements’ summary of the above session.
One day I’d like to actually see Mark Hancock present! This year it was about accents:
Laura Patsko’s session on how to give feedback on learners’ pronunciation was one I was sad to miss, but luckily Cambridge recorded it, so I’ll be able to catch up. Here are some tweets to give you a flavour of it: