Pronunciation is a motor skill. To be able to produce the sounds of a language correctly we need to work on the movements of the mouth (and so on). However, that’s not all pronunciation is. We still need to develop muscular memory and automaticity in order to achieve fluency.
ie. repeating things makes it easier for your mouth to make the right sounds when you need them, and having phrases that you can utter without thinking makes you more fluent.
Put even more simply, practice makes perfect and repetition helps with memorisation. This is a case for using drills, shadowing or other techniques. Although looking back at what I’ve just written, it’s more about improving fluency and memorisation than it is about pronunciation… Food for thought!
Making drilling more fun
Many teachers are put off from using drills because the “listen-and-repeat-after-me” (or “after the CD”) kind of drill is pretty boring. There are several ways to liven up drilling.
Here I’m going to describe one choral drill that focuses on sentence stress, using a stamp of the foot on the stressed word (or words). Surprisingly I didn’t find it by “drill with stamping” or any similar name by googling, so perhaps it has another name? It’s also good as a Question-Answer drill which makes it more interactive and a bit of “substitution drilling” can be added in, as you’ll see later.
I’ve tried this with low-level, shy Vietnamese adults and tweens but I haven’t (yet) tried it with older teens. It’s a bit physical, gets students out of their seats and the students I’ve done it with have enjoyed it.
Choral drill with stamped stress
I’ll illustrate it with an example for ease:
A: “What time is it?”
B: “It’s __ o’clock”
Setup and Modelling
Put the students standing up in two lines facing each other (not too close to each other or it would be a bit intimidating)
Model the phrase and stamp your foot (stepping slightly forward as you do) to indicate the stress as you say it. “What TIME [stamp] is it?”
Drill the whole class on the phrase, including the stamping on the stressed words.
Model and drill various answers, eg. “It’s 4’o’clock”, “It’s 10’o’clock”, etc.
Chorally drill half the class (ie. one line, let’s call it Line A) Ss on the question: “What time is it?”. The other students, Line B, remain silent.
Drill the other students, B, on an answer or two: “It’s 4’o’clock”. Line A remain silent.
Explain or model that when you count down 3, 2, 1, then A will say their line and B immediately answer with theirs.
Call out a variable to change, eg “ten” and they say it for “ten o’clock”.
After a few drills with A asking B for the time, you then swap them round so Line B are now asking “What time is it?” and A answer. Top tip: I’d advise quickly re-drilling each group before you do a 3,2,1 to start it off.
It can take a little while for students to get the timing right (ie. for students get that it’s 3,2,1 – speak).
While there is drilling, after the initial setup, it’s more student-centred and less a case of “repeat after me” once you get onto the main stage of the drill.
Considering a silent approach to pronunciation, you could probably remove the listen-and-repeat “What time is it?” drill with humming the rhythm: də–Də-də-də?. I think I have done this in the past (I first learnt this drill in 2012).
I’ve also done a little bit of individual drilling and/or given feedback while students are stood in the line.
Students do tend to find it quite fun. It’s potentially a drill to add to your repertoire, to use occasionally. It might help boost students’ confidence and fluency before doing a mingle or whatever you plan do next.
Over to you: What variations do you add so your drills aren’t just listen-and-repeat? Or do you avoid drills entirely? (If so, I’m curious why!)
PS. It’s hard explaining this online, and I may have missed something, so do ask if anything’s not clear!
Connected speech has many elements. I’m assuming you are at least a little familiar with these features. If you’ve never heard of terms like linking, assimilation and elision, Rachael Roberts has a great blog post which introduces the main players in connected speech.
Perhaps you’re wondering what’s the point. Are these words just jargon to learn for the Delta exam? Is there any application to the classroom? Well, of course students don’t really need to learn terminology like elision, but connected speech is most definitely relevant to the classroom. Did the title of this article give it away? 😉
Why learners need to learn about connected speech
They don’t necessarily need to start speaking like that themselves. But if they are not aware of the changes that occur in fast, natural speech then chances are that learners’ listening ability is going to suffer.
The stream of speech
Natural, fast speech is “full of simplifications and reductions” (Hancock 2013) in terms of pronunciation. Phonemes can change or disappear in connected speech because doing this makes speech faster and easier for us. Even the position of words in the sentence can change how they sound, and they can sound really rather different. Unrecognisable even, to someone who’s not aware of these changes.
As expert listeners ourselves, we are used to decoding the “stream of speech” we hear. We map the reductions and simplifications we hear to words and phrases we know, filling in gaps that weren’t even there. We hear traces of a sound and think we’ve heard the whole thing (Cauldwell 2013). And we do it automatically.
But most of our learners don’t. Is that surprising? Chances are, changes in fast, natural speech are different in their own language. Chances are, they haven’t listened to enough English to be anywhere near an L1 style of acquisition. Chances are, they’re pronouncing English clearly themselves: “What are you going to do?” instead of “What’re ya gonna do?”. Chances are, no-one’s ever taught them about these changes.
No wonder they can’t understand. 🙁
There are many features of connected speech. Where to start?
Now, that doesn’t mean you should go into class tomorrow and spend the whole lesson teaching learners about all the features connected speech. It may well overwhelm them (or you, or both).
I’ve been doing this for three years now and I go through the features one or two at a time (as per Kenworthy 1987) as they naturally relate to something in class. I focus on the ones that get students the most mileage. As a result, I rarely teach some of the features like intrusion.
So, which features of connected speech get students the most mileage? Which do they really need to know, asap?
Starting with the schwa
It’s no secret both I and my students are familiar with the schwa*. This is generally the first element of connected speech I teach my students.
Not only is it more straightforward than some other elements of connected speech, many of the other features (like linking and assimilation) will often include a schwa. For example, in fast speech a lot of often has linking and two schwas: /ə lɒ təv/.
The schwa comes up all the time, so you can introduce it in pretty much any lesson. Often the first time occurrence of the schwa I draw attention to is prepositions like of or to. I don’t phonemically transcribe the whole word, but I do usually board the /ə/ phoneme.
After eliciting the weak forms of some other prepositions (eg. from), I then elicit other function words like auxiliary verbs. Depending on their background, many students will have had some exposure to the schwa. If not, I still get them to produce it. Eg. “How do you think we say was /wɒz/?” (wəz)
Once students are used to the schwa (and sometimes they’re familiar with the sound before I start teaching them, so this can be in the same lesson), I often move on to assimilation. This is not always the second feature I teach, but lately it has been.
I start with reductions that happen in very common question forms that students are going to hear all the time. If you want to be precise, I’m talking about coalescent assimilation here. But we can just call it assimilation.
For the last two groups I introduced it with, these were the two questions I used and the context they came up in to show you how you can link it in with a lesson.
Do you like…?
/duː juː/ becomes /ʤə/ (or a softer /djə/ which isn’t technically assimilation)
While teaching in the UK this summer, my high pre-intermediate students were going to go out and survey local people. The students were a little nervous about starting conversations with locals so we spent part of the lesson preparing for this, including practising the pronunciation in order to build their confidence.
One of the questions was “Do you like living here?”. I seized it as a chance to introduce students to assimilation. It blew their minds. They laughed a few times as they practised trying to say it in the classroom. This is good, we’re not aiming for perfection! Make practising connected speech light-hearted.
I don’t know how they pronounced the “do you” question out in public, but it doesn’t really matter. The point of teaching connected speech is more for listening, so they can understand when other people are using it.
What’re you gonna do this weekend?
reduced to /wɒʧə gənə duː/ (or /gʌnə/)
I wanted to make small talk with my low pre-intermediate students during the Delta, particularly as I often taught on a Thursday or Friday. This question was problematic – they didn’t understand me unless I slowed my speech right down.
One day I was only scheduled to teach them for 30 minutes. Hello, assimilation! We spent the whole time practising this structure, focusing on both pronunciation and grammatical accuracy and finishing off with some small talk (eg. “What’re you gonna do for Christmas?”).
And then what?
It’s important not to focus on connected speech in one lesson and then never speak of it again. Exploit the schwa and assimilation when they come up in following lessons, keep raising students’ awareness of these features of connected speech and encourage them to notice them for themselves.
Recycle the questions that you used to introduce assimilation. This was the main reason I chose “What’re you gonna do this weekend?” – for the small talk which could be included at the beginning or end of lessons!
Draw attention to schwas and assimilation that come up in class. Over the course of 3-4 lessons my pre-int students became reasonably familiar with this kind of assimilation.
Exploit the listenings you do in class. After comprehension work, you could exploit the audio and/or the transcript. Encourage learners to notice schwas and/or assimilation. Remember to only focus on one or two features of connected speech at a time (Kenworthy 1987). The lower the level, the more likely I’m just going to focus on one feature at any one time.
Predict the pronunciation
When presenting new language, elicit from students where they expect schwas and assimilation.
You could exploit written language – from instructions on the page in a textbook to an excerpt from a reading. Get students to tell you what schwas or assimilation they’d expect.
Elisabeth, I still have questions
You may have some questions about the above. I’ve tried to anticipate some of them here, but let me know if you have any more! I love questions (no joke).
Q. But you’re teaching connected speech to low levels!
A. Yes. In my opinion, the sooner the better. For adult learners, I would introduce the schwa in one of my first lessons at any level, beginner to advanced. Assimilation I’d teach from false beginner to advanced. I teach quite a lot of Elementary classes and I always introduce these two features early on. Even my Primary YL students know about the schwa sound!
At the moment I’m coaching a C2 level learner who’s not satisfied with their listening ability. Cauldwell at IATEFL 2017 also mentioned a student who passed CPE but had accepted they ‘couldn’t understand native speakers’. This saddens me. I’d like for my students to develop their listening throughout their learning journey so they don’t end up in that position!
Q. But why teach students to pronounce connected speech if they don’t need to start speaking like that?
Cauldwell (2013) says students can “improve their decoding skills by repeating, changing, manipulating and interacting in a variety of ways with the sound substance”. Basically by playing around with the sound changes themselves, they become better able to ‘decode’ or understand these changes when they hear them.
For example, if you tell students “going to” is pronounced gonna, they’ll probably accept it but may well forget it. If you spend 10 minutes getting students to practice saying gonna in a variety of sentences, they may have forgotten it by the next lesson. They probably won’t start using gonna themselves. But they’re probably more likely to recall it the next time they heargonna. When they hear gonna and understand… That’s what we’re aiming for.
Q. Any other tips for starting out?
A. Particularly for teaching low levels, I like to joke that we’re lazy and we like to shorten things. I then gesture with my hands to indicate shorten. Then, when practising, if learners are producing the long, “clear speech” form, I use the gesture to encourage them to self-correct to the messy connected speech version.
Students do sometimes ask me why we use connected speech. I usually simplify it to ‘because it’s easier for us to pronounce’. They can look a bit doubtful at this point, so if you’re aware of any connected speech in their L1, it can help to refer to it. I usually then reassure them again they don’t have to speak like that, but native speakers do so they’ve got to be aware.
Over to you: Do you teach connected speech to improve your students’ listening? Where do you start? If you don’t teach connected speech, do you have any questions that I haven’t covered above?
* I’m starting to fall out of love with the schwa. It’s an ongoing process, but I’m sure we’ll still be friends at the end of it.
From the very name of the approach “The Silent Way” I dismissed it as I believe input (or exposure) is one of the three conditions necessary for successful language learning.
This exposure to what I’d classify as “comprehensible input” (from both the teacher and my classmates) is certainly one of the main reasons I attend group language classes myself, as a language learner. So I don’t like the idea of the teacher not being a source of input and listening practice for learners.
As a teacher, my classes are very student-centred and too much TTT has never really been a problem for me. However I do of course talk to my students.
How can “silent” and “pronunciation” go together?
You may be aware of Adrian Underhill’s way of introducing pronunciation. If you aren’t, watch his video. When I first saw him present this, I started to understand pronunciation in an entirely different way. I learnt how sounds were formed and how to show my students how to move their mouths into the right position for sounds they were struggling with.
Pronunciation is a motor skill. Pronunciation experts like Messum, whose webinar I watched as part of IATEFL #webconf17, say we don’t actually learn pronunciation by hearing and copying but that we need to learn how the mouth, tongue, lips and even body are used to produce the sounds of English.
Let’s revisit Underhill’s approach with this in mind.
You may or may not have realised he never says the sounds as he’s teaching them. He shows students how to position their mouths and the students themselves actually produce the sounds. He gives them feedback on their performance, whether it’s good or if not, feedback on how to improve. This (in my understanding) is teaching pronunciation the Silent Way.
In the classroom
I was sort of aware of this before, but perhaps not fully aware of the implications. I have definitely at some point said things like “learners can’t say what they can’t hear” (and vice versa). But I’m no longer sure that’s entirely true. If learners position their mouths correctly, won’t they produce more or less the right sound?
Of course, it is not easy to get learners to position their mouths correctly.
I’ve tried. I’ve used Underhill’s way of teaching the phonemic chart to three different groups: speakers of Vietnamese in 2015, speakers of Korean (and one Arabic) in 2016 and speakers of Spanish in 2017. I did initially try not to say the sounds myself but I have to admit I did sometimes say them. Habit! The most recent time I taught the chart, I definitely caught myself modelling too much and was aware of it at the time. However I’m pretty sure I was just thinking “oops, stop modelling” rather than considering the reasons behind it and why it’s so important to show rather than model.
Now I understand how the Silent Way can be used for developing pronunciation as a motor skill, the next time I teach phonemic sounds (or the chart itself), I’m going to be doubly conscious of this.
Over to you: How do you teach pronunciation, particularly the phonemic sounds?
References and Further Reading
This post was inspired by November’s IATEFL Web Conference 2017, realisations from Messum’s excellent session on pronunciation and reading I did afterwards.
to – one of those words that, most of the time, is unstressed and uses the schwa /ə/, right?
That’s what I thought until one of my Delta Module 2 input sessions this week when my mind was thoroughly blown. It turns out it’s dependent on what comes next.
Let’s take an example with telling the time, as last year both me and my Primary students had fun when drilling weak forms.
It’s ten to two.
It’s ten to eight.
In the first example, all is as expected: to is unstressed /tə/.
But in the second example other phonological rules come into play, don’t they? Schwa /ə/ is a vowel and is followed by eight which starts with the diphthong /eɪ/.
What happens with two consecutive vowels in connected speech? Why I’m glad you asked – what we get is an intrusive /j/, /r/ or /w/ depending on the sounds involved. Here it’s /w/.
In #2, I think the schwa is still there (though I am no phonology expert and could be wrong). I think that most of the time, there is still the intrusive /w/ which means it comes out as /təweɪt/. Most of the time. I’ve been saying it to myself and sometimes I’ve not heard or felt my mouth move to the /w/ position at all. But in another example such as to eat /təwiːt/, the /ə/ and /w/ are clearly there to my ear (and pronounced differently to tweet /twiːt/).
So, what does this mean for the classroom? The way we drill the sentence (or phrase) changes. It’s not quite as simple as /tə/ + /eɪt/. We naturally introduce the /w/ when speaking at full speed, but is it always there when drilling? I’m not sure. I think it’s entirely possible I’ve oversimplified it when teaching.
Going forward, I’ll be watching out for this!
Over to you: Have you thought about this aspect of the schwa before? What have you learned or been surprised by in your teaching of phonology?