George Loetter has been involved in the ESL industry in Africa and Asia for the past 12 years. Currently, he is an Assistant Professor at Kansai Gaidai University in Osaka, Japan, where, as always, he continues to learn, teach, and create, while keeping things as simple and practical as possible. Email: mrloetter@gmail.com

If you've taught ESL at a variety of institutes, covering different levels and age groups, you might have noticed that the same pronunciation points appear in textbooks across the board; from elementary phonics to university second year courses, to in-house adult learning materials. Some teachers tackle these parts with isolated drills, some use minimal pairs, others provide descriptions of articulators, and others use mirrors and exaggerated examples, some use a combination of all of these - and there are teachers who just skip these sections entirely. 

Regardless of the approach to pronunciation - why is it that there's an eternal attempt at pronunciation practice for phonemes, intonation patterns, and possible blends, that have appeared in student's textbooks consistently for going on 12 years (depending on where we meet them within their journey through the education system) - and should we be doing classroom pronunciation practice each time these sections appear in the various curricula we are required to teach? 

 

The why and the should 

One simple answer to why this happens is this: A lack of sufficient input. A slightly more detailed reason: A lack of sufficient input, a lack of motivation for students to independently progress beyond basic requirements, the resulting pressure from the idea that speaking quickly makes one "sound native", general low education standards, and all this while approaching language as a collection of isolated parts that, if practiced separately, will lead to natural pronunciation and understanding when all the parts are later combined (in case it isn't clear, I'm not really on board with this Frankenstein approach). 

An answer to the question of if we should keep taking time to specifically focus on these repeated elements of pronunciation during class? I don't have one, because that's up to you and what you feel is best for your students. When teachers fully believe in their techniques (and the background principles that motivate them), magic happens - so it's not my place to decide what works and what doesn't for you, but I can share my personal answer to the question, what I do (and have done) and why I do it. 

 

My take on pronunciation 

The "why" linked to my "what" has been touched on already, as it is made up of counterpoints to the reasons offered above: 

Pronunciation is all about getting sufficient input so that the ear can be trained, an independent attempt to get this input, slowing down and putting the "sounding native" pressure on the backburner, and allowing the brain to use the received input, in all it's cohesive, rich glory, so that the articulators naturally produce the sounds that accompany acquisition once the ear has been trained and utterances are good to go. Isolated pronunciation drills, and an understanding of articulators, are not a requirement for any of these points.

Of course, we must operate within a classroom and with the students we get, and it's not realistic to expect this process to run its course with students who are taking mandatory classes once or twice a week - but we can use these principles when focused on skills training (in my opinion, the most important aspect of ESL teaching) and make our classrooms a reflection of the techniques that students can use when moving through these processes as independent language learners. 

Do I take whole groups through sessions of isolated pronunciation practice, explaining articulators and different elements of pronunciation that I know they've come across countless times before in a listen and repeat manner? No. 

Specific pronunciation training is only necessary in cases where someone has a specific problem and needs one-on-one instruction to correct it. It’s in the realm of speech therapy rather than something that should take up a lot of time in normal ESL classes. Perhaps it would apply in a group setting if you're lucky enough to be in a class that has been so perfectly levelled that you're teaching a group who all have the exact same issue and have all requested some speech therapy to pinpoint and correct a specific sound, which is not a usual situation. This is not the same as focusing on a common, assumed error among a cultural group, such as the notorious "l/r" differentiation issues seen among Japanese and Korean speakers. Not all students will need this, especially year after year in a classroom setting. 

 

Some of them are really good already, right?

The students with an independent interest in language, who consume L2 media and eventually speak after moving through natural stages of acquisition, can produce these sounds without any issues. We know who they are when they appear in our classes. Try asking them to explain how exactly to produce the sounds they make that their peers struggle with - they will barely know, or perhaps fumble through what they think a teacher once said when trying to explain it. Their ears have been trained, their brains and articulators adjusted automatically, they gradually produced when they were ready, or were in situations where they were required to produce consistently over longer periods of time, and natural acquisition did its job (which is also the case for those who got there through a lot of independent repetition motivated purely by a desire to get high grades). 

 

Should the rest eat toast?

Doing 10 minutes of isolated drills in open class while pointing at diagrams and emphasizing mouth movements will do as much for the students who aren't used to hearing and saying a certain sound (beyond the eternal phonics treadmill they have been doing in mandatory classes) as asking them to eat more toast for breakfast. The same goes for quickly correcting a struggling student until they produce what sounds correct and then moving on - you might as well ask them to add some Bovril to that toast. It works just as well. 

 

So, what do I do in class then?

Instead of isolating pronunciation practice and explicitly learning the manner and places of articulation, I prefer to use class time to focus on the ability to hear the target sounds within natural speech, and on doing activities that serve as demonstrations of what students can do independently to improve. You can’t produce a sound if you can’t correctly hear the sound in the first place (particularly if you can’t hear that sound in context) and hearing all the sounds of a language (and getting used to intonation patterns and emphases) takes a lot of repetition of natural language. Once your ear is trained and your brain has made sense of sounds naturally, then it becomes easier to practice and reproduce, which should also be done in context, without separating single sounds. 

Distilling this process so that it fits into an ESL class means focusing on lots of listening input, with as much comprehensibility as possible, then giving the students opportunities to practice output. Ideally, listening and comprehension would dominate until students are ready to speak, without forcing premature output, but the limitations and time restrictions of a classroom environment necessitate shortening the process, adding some pushed output, and including skills training to counter deficiencies. I find the best activities for this to be listening and reading at the same time, dictation exercises, shadowing, and dialogue practice with known dialogues, which are all also good starting points for designing dynamic tasks, interesting activities and eventually more open conversation practice. In this article I just want to focus on listening and pronunciation however, and for that, these are some example activities that utilize the above principles:

 

Practical examples using texts

Choose a short reading passage that is useful or interesting (ideally both), identify the words and clusters containing the sounds to be focused on (more than one sound can be chosen). Create a cloze passage using these areas as gaps and get students to listen to the audio of the passage while filling in the blanks. This will help students to focus carefully on the missing words and begin picking up the familiar pattern of the sounds that are the focus for the day, particularly since the passage will need to be played multiple times. A variation on this is to give students the full passage, but with each word or phrase containing the target sounds for the day replaced (ideally with a minimal pair or something similar). Students then listen carefully, circling and correcting the passage when they spot a difference between the audio and the text.

If the text doesn’t come with audio - read it yourself. You can also show students how to read and listen to any text that does not come with audio by using Microsoft Edge. Save the text to PDF, open it in Edge, and use the read aloud feature. Edge has excellent AI text-to-speech technology, which is head and shoulders above the rest in terms of natural speech. It also has the advantage of being able to select different dialects, which can make for some interesting classroom listening exercises, and if you’re using text from a website, there’s a translation feature that allows certain words to switch to the L1 without translating the entire passage. It’s even possible to colour code different parts of speech. Audio made specifically for a text, or a teacher reading, comes first - but Edge is an excellent tool after this if needed, and serves as skills training as you will be demonstrating what students can do themselves at home. 

Students can then pair up and confirm answers by saying each answer to their partner, reading the sentence or phrase, not individual words. 

Finally, in open class, answers can be confirmed, with the whole class repeating the phrase said out loud by the instructor after each answer is volunteered by a student. Therefore, non-isolated drilling is used, but with new content each time, and the target sounds for the day mixed in. This provides more repetition of what has been heard multiple times and satisfies the need to add more output practice. 

 

Practical examples using dialogues

Dialogues can be sourced from the textbook, TV and movie clips online, or can be made by the instructor, and can be used in the same way as texts, making cloze (or spot the difference) listening exercises and using the sequence outlined above. What’s more, they are also a good resource for practicing shadowing, and adding more levels of pushed output. 

Shadowing is a technique used by many independent language learners, which requires what the name indicates - utterances are shadowed. After each line or half line is heard, the learner immediately repeats what was said. They do this repeatedly until they can comfortably say the line right after it has been said, saying it sooner and sooner with the eventual goal of saying the line at the same time (or very close to the same time) as the speaker, also matching the intonation. This should be done after the dialogue has been seen and understood, and without the text. It should purely be listening then mimicking sounds and intonation. The aim is to speed things up in a controlled setting, allowing the brain to grow accustomed to hearing and producing sounds in context, without slowly isolating any parts. Ideally, the instructor provides the audio so that the pace can be modified as needed; but good quality audio along with pausing and repeating will do the job just as well, while providing an example of how to do this at home for the student. 

I would recommend using short dialogues in class, doing one line at a time, repeating it a few times as the class gradually shadows closer and closer, and then doing the whole dialogue. Quick, interesting dialogues are best to avoid classroom fatigue, but motivated students doing this at home can use any length of dialogue they choose. 

After this, students can practice the dialogue with each other in pairs - first reading as a refresher, then taking turns to read while their partner says their lines without reading (with the reader offering clues, prompts and support). Finally, students try to say the dialogue with no text at all (encouraging mistakes and deviations, as long as the general meaning remains, is a good idea for these types of pushed output segments). 

 

“Native speakers speak fast!”

While teaching English majors at a university in China, I came across a common problem that I believe is rooted in the idea that good English “sounds native.” Students wanted to sound British or American, believed that good English was fast English because “native speakers speak fast”, and would try to blend certain sounds correctly through dissecting these fragments of speech and practicing them in isolation, before enough input had been gained to identify sounds clearly and reproduce them naturally. Motivation for this came from classes they were taking simultaneously, teaching them how natural, everyday utterances sound when blended - and how to achieve the same effect (an example of this is how “What’s the time?” can sound more like “Whatsa time?” when spoken quickly by those fluent in English). 

There are course books, textbook chapters, teacher training modules dedicated to this aspect of speech, and how to teach students to blend correctly so that they can sound natural. Even more so than isolating phonemes and practicing them inorganically, I see this as adding complication and stress - slowing down the acquisition process rather than aiding it. Even students without the basics of decent pronunciation yet acquired would pull these utterances out at random, almost like a party trick to demonstrate superior pronunciation and language skills. The result of focusing on these signposts of what English “should” sound like versus gaining input and pronouncing each sound clearly after it has become familiar by the ear and recognized in patterns by the brain, is that students would speak at their normal level most of the time, would sometimes insert faster phrases that didn’t match that level (and were sometimes inserted in awkwardly), and some would try to speak quickly and blend sounds before quite acquiring a good pronunciation in those same words when speaking slowly. Even worse, it would put pressure on many students who could not express ideas quickly and fluently, particularly when comparing themselves to peers producing more “fluent” speech.

The ironic part of all this emphasis on blending sounds to sound natural is that slow, well-pronounced speech is the fuel for natural, fast, blended speech - no real training required. Say the following sentence, pronouncing every word carefully and clearly: “What are you eating for lunch?” Now, say it again, but faster. Then again, even faster. Say it as fast as you can. Was it even possible to avoid saying something that sounds more like “Whadaaya eeding fulunch?” (with slight variations depending on accent and dialect). We are not trained to change our speech and blend it in the right way - it happens naturally when we say what we know, and say it fast. In the same way, ear training (combined with reading initially to assist comprehension), allows this to be identified after lots of natural listening input is acquired - no need for special training, it will be observed and identified naturally as a learner gains more input. 

The solution for my students at the time then, was to hit the brakes. I would ask students to slow down and pronounce every single word carefully, even if it sounded unnatural to them (it never seems to sound as slow as people think it does when coming from their own mouths though - record yourself and see for yourself, you can speak at quite a healthy pace even when pronouncing every word clearly!). I encouraged them to do this for a few weeks, and those who were more motivated and asked for assistance outside of the class would talk to me pronouncing every word as clearly as possible, following the advice. Some did it to show they were giving it a try, some just to joke around - but many did try regardless, and what happened when they were asked to “let loose” and could speed up certain words and phrases they were comfortable with? More natural speech and blended sounds that would appear of their own accord as if my magic. I didn’t do a full-on scientific study to measure the difference, but I do believe this is a good way to get students to slow down, focus on the basics, speak and speed up when ready - and natural speech will result. What’s more - getting students to practice this for weeks on their own to bring home the point isn’t necessary; pacing the dialogue practice (or any sentences used in class) so that students gradually say sentences faster and faster, will offer a demonstration of this concept at work. 

On a final note, there are cases where some words don’t blend as easily and are taken as almost new, separate words, such as how “What are you going to do tomorrow?” sounds like “Whataya gonna do tomorrow?” for many people, yet some always say “Whataya going to do tomorrow?” and some say “gonna” even when speaking slowly. In these cases, “gonna” might not result, but I’m of the opinion that clear, well-spoken speech (regardless of accent or dialect) sounds intelligent and is a good aim for language learners. Common variations, many bordering on slang, can be assimilated in time through choice depending on the content the students choos to immerse in as independent language learners, and in how they choose to interact and speak later on. 

A counter I’ve heard to this is that the machine-gun or monotone tendencies of some speakers are corrected by learning how to properly blend sounds in English - but this is not a problem of blending. This is a problem of stress and intonation. These areas are more useful to teach, and if a focused, isolated pronunciation session was required in each class, I would rather use it on improving stress and intonation. Even so, it is included by default when shadowing, and when training the ear - those students who get a lot of input and practice will naturally improve their pronunciation in all areas. 

 

A note on comprehensibility 

One of the elements of providing students with texts, dialogues and accompanying audio as a form of input is to find or create texts that are appropriate for the level of the student (I’m sure anyone who has read this far is familiar with the term “comprehensible input”, or “i+1”). Although it still trains the ear in terms of intonation, we don’t want students listening to gibberish, as it’s hard to focus on listening to sounds and remain motivated when you don’t know what you’re listening to (as noted - useful, interesting and ideally both is what we want). 

However, when choosing a text, it’s better to land on one that is too difficult rather than too easy, as the day-to-day of independent language learning, if using natural language in context, does not generally involve tailored texts - and part of classroom skills training can be to prepare with a text before doing the above activities. A common, almost instinctual (perhaps habitual?) approach is to pre-teach vocabulary, which, when sticking to the motivation for this article, also begs the question: do all students need to be taught the exact same vocabulary within a target text? They likely don’t, and this issue goes away if students are given an opportunity to identify their own unknown vocabulary and, through skills training, learn what to do with unfamiliar words. 

Students can be given the text and some time to read it, identify unknown words and phrases, and use dictionaries and translation apps to look up what these words and phrases mean - particularly those that cannot be figured out from context. This mimics what an independent language learner would do outside of the classroom. They can then add these words and phrases into a spaced repetition system, such as Anki, the Goldlist Method, a deck of flashcards (the most unwieldy option) or any variation you choose, which will require some initial training in class - both in how to use the chosen system, and the best way to make flash cards (or entries in the Goldlist Method, which use the same concept). This can be extended into a weekly homework activity, where students are required to complete their lists with any interesting content they find away from the classroom.

I like to use a simplified version of the Goldlist Method (one list a week, distilled into new lists of the same size every second week), using intervals that align with weekly classes throughout the semester, as the creation and recall processes can be done in class (with all students on the same task, using self-selected vocabulary, making this suitable for mixed level classes), turning the class into a demonstration of how to do this independently. It’s also a good idea to give students the option of using Anki, which is powerful SRS software that is very popular among the independent language learning community and is available for free on Android, Windows and Mac. The method you choose to structure entries depends on your philosophy of language learning. Some options for creating flashcards and list entries (“back” also refers to “opposite side” if using lists):

• Write the full sentence in the L2 on the front, with an L1 translation on the back (lower stress, just focuses on recognition). This is how the real Goldlist Method should be done - a relaxed exercise in recognition, not a test (which is what recall can feel like).
• Write the full sentence in the L2 on the front with the target vocabulary missing, and the L2 answer and L1 translation on the back (lower stress, blends recognition and recall). 
• Write only the L2 word on the front, and the L1 translation and L2 sample sentence on the back (lower stress, just focuses on recognition, removes remembering meaning of sentence rather than actual target word)
• Write only the L2 word on the front, and the L2 sentence as a clue on the back (this forces a break of reliance on the L1, and the effort of looking up the meaning again, if something is forgotten, improves future recall). 
• Write the L1 word or sentence on the front, and the L2 translation on the back (better for more intensive study, focuses entirely on recall, increases stress). 

The best option is to keep things simple and relaxed, using the first method for classes. The other methods can gradually be taught to students as the semester progresses, and they can use these outside of the classroom as they see fit. 

 

From isolated pronunciation practice to a recipe for language learning 

A quick recap. Students receive an interesting or useful (ideally both) text. They use this to identify the words they don’t know, discover the meaning and add new vocabulary to an SRS system. Then input is maximized, and ear training begins, by repeatedly listening to and reading texts at the same time (with focusing exercises, particularly on target phonemes), after which shadowing takes place, where students produce the sentences, pronunciation and intonation patterns they have come to understand and have already heard repeatedly. Finally, more open discussion and activities can take place (which moves beyond pronunciation training and delves into activities beyond the aim of this article). 

Put another way, isolated pronunciation practice has shifted towards a process that gives students all the tools they need to independently improve their language skills. When not in the classroom, they should read texts that are useful and interesting to them, look up words they don’t know, use a system that aids in long-term recognition and recall (SRS), get lots of listening input (ideally by reading and listening at the same time - at least initially), shadow to improve pronunciation, and speaking should happen when they can identify the sounds of natural speech and are ready to talk. 

The only missing ingredient is resources. Usually, a course book will be available and required texts and audio will come from there, but it’s always a good idea to use content from interesting resources that the students can access themselves (again, adding skills training to the classroom). Edge was already mentioned due to its versatility and excellent toolset (opening up any online page as a resource for language learning), and there is an abundance of materials available online for free. For example, Ello is a fantastic resource for natural dialogues with transcriptions, VoA offers interesting news stories with audio, and the Language Reactor add-on for Chrome or Edge turns Netflix or YouTube into a language learning tool. 

With that, the Frankenstein monster is gone, the classroom becomes a reflection of the language acquisition process, and of how independent language learners approach languages, and good pronunciation skills result as a natural by-product.

 

Please check the Pilgrims f2f courses at Pilgrims website.

Please check the Pilgrims online courses at Pilgrims website.