Priorities in Pronunciation Teaching

The teaching of pronunciation is often a challenge for teachers. This may be due to the fact that there is so much to accomplish – so many phonemes to deal with, so many intonation patterns, connected speech features – that we teachers often feel overwhelmed, and perhaps even insecure of our own abilities to give our learners what they need to develop their pronunciation. We have the aid of our course books, which highlight some pronunciation features, focusing on phonemes and other segmental and suprasegmental features, and providing practice exercises and phonemic charts, which we use, when they come up, as suggested in the books. Whenever we teach new language items, we make sure students are exposed to and get enough practice of the correct pronunciation of words and phrases. Also, in our everyday classroom practice, we know that correcting students’ mistakes is very important to help them achieve an acceptable pronunciation. This is often what many of us do at any sign of error. However, we might simply become so accustomed to hearing recurrent errors that we just do not notice them any more.

But the most important aspect of pronunciation teaching may not be how we teach it, how much we teach it, or even what we teach. Perhaps the crucial question to consider is why we teach pronunciation. Only by asking ourselves this question, and finding a satisfactory answer to it, can we really have a clear picture of the importance of pronunciation in our students’ lives, and begin to think of answers to the other ones.

In order to focus on the question of why it is important to teach pronunciation, let us have a look at the following example sentences:

  • Doo you wanty sã coff? Wivvy mewk or widouty mewk?
  • Dat sin yooman works in an offs. I don’t sink she’s he-tir-ed.
  • Dis is my cless-HÕÕ. Doo you likey ITTY?
  • We watch-ed Hobbin Hood yesterday. I was heally happ!
  • In the intermedi-ATE course we have a lot of home-WORK.
  • Bew is at the entrance how in the cown-trie cloob.

You probably found these written sentences hard to understand when you first glanced at them on paper. But I am quite sure that as soon as you read them out loud to yourself and gave them sound, you certainly recognized them immediately! (If not, take a look at the footnote for their real English versions.1) You must have heard – perhaps too often – words and groups of words sounding like that in your classroom. And it probably bothers you to hear your students speaking like that. But what can we do when we hear such sentences? Can we just ignore the errors because this is the way some Brazilians will pronounce English, and there is nothing we can do about that? Should we try and correct all possible errors, and make the student repeat the correct pronunciation until we are satisfied (or until we give up!)? Or should we use them as an opportunity to analyse these errors and interpret our own reactions to them? Why do these utterances bother us? Is it simply because they ‘sound bad’ and ‘hurt our ears’? Is it simply because they are incorrect? Perhaps the real reason why we feel students should not speak like that is the fact that we know – even if only on a subconscious level – that this sort of pronunciation is likely to affect the speaker’s intelligibility when communicating in English.

1Do you want some coffee? With milk or without milk?
That thin woman works in an office. I don’t think she’s retired.
This is my classroom. Do you like it?
We watched Robin Hood yesterday. I was really happy!
In the intermediate course we have a lot of homework.
Bill is at the entrance hall in the country club.


Intelligibility – a reliable guide

Being intelligible – i.e., able to make one’s meanings and intentions clear to a listener, according to a simple definition by Brazil (1994: 2) – is obviously essential, but it will not be guaranteed by correct grammar and appropriate choice of words only. If the speech flow is not produced in such a way that it can be recognized by a listener, then the message will not be conveyed. Also, if one’s expectations of what words and phrases should sound like is inaccurate, they are not likely to understand what is said to them. This is what makes the teaching of pronunciation so important.

But who should our students be intelligible to? When they utter sentences such as the examples above, can they understand each other? Most probably yes, because many of them share the same problems. Can you understand them? I am sure you can, as you must be more than familiar with common Brazilian pronunciation difficulties.

But will they be understood by native speakers of English? Will they be understood by speakers of other languages? And will they be able to understand a correct pronunciation when they hear it?

The aim of our teaching is basically to enable our learners to use English as a tool to communicate with people in a world where English has become the most widely used international language – the lingua franca. So it is crucial that their pronunciation is intelligible to a wide variety of interlocutors. However, that does not mean to say that we should aim for a perfect, native-like pronunciation. According to Walker (2002: 9) “whilst it is perfectly legitimate for a student to aspire to a native speaker accent, it is surely wrong for a teacher, explicitly or otherwise, to push students to feel that anything other than this is an imperfection.” Another important aspect to consider is the fact that Brazilian learners will probably need English to communicate with speakers of other languages a lot more often than with native speakers (consider, for example, the recent expansion of trade between Brazil and African and Asian countries, and the rising importance of the European Community). Gnutzmann (2000: 357) reminds us that “it has been estimated that about 80 per cent of verbal exchanges in which English is used as a second or foreign language do not involve native speakers of English (Beneke 1991)”.

If this is the case, we should concentrate our efforts on teaching our learners what is really essential to make themselves understood and to understand what is said to them. In order to do that, as suggested by Jenkins (2000:104), we must become aware of which pronunciation problems really do affect their intelligibility and which do not, rather than ‘shoot in all directions’ aimlessly after a perfect pronunciation of all phonemes, clusters, intonation patterns, etc. In most cases this perfection proves impossible to achieve, and the quest for it tends to bring a lot of frustration to both teachers and students, and does not provide any focus or aim.

If international intelligibility is what we should aim for, instead of aiming for native-like pronunciation in all areas, we should learn to prioritise our pronunciation teaching. Only by doing that will we be able to establish which areas to tackle more intensely, which to be only relatively concerned about, and which to pay less attention to when teaching and correcting pronunciation, so as to maximise our time and effort.

Learning what our learners need most

The example sentences we looked at above can help us a lot in this prioritisation if we analyse them in the light of following questions:

  • Which typical Brazilian pronunciation errors are students making in them?
  • Which of these errors are more or less problematic as far as international intelligibility is concerned?
  • How should we deal with each of them?

The answer to the first question is so overwhelming that it would probably take us a few hundred words to list and analyse them all. Among the ones Brazilian teachers traditionally pick up on immediately are the mispronunciation of the two TH sounds (*I sink, *dat), the addition of an /i/ sound after final consonants (*likey, *itty), the “Brazilianized” pronunciation of loan words like country club and hall, and, more recently, the mispronunciation of the Past Tense suffix ~ed. But other errors involving different segmental and suprasegmental features – which are sometimes not even noticed, and frequently not perceived as very serious – can also be important.2 That can only be evaluated after we answer the second question.

Finding a definite answer to the second question is a lot more difficult. How can we have an accurate idea of which errors may hinder comprehension in an international context? Here we can count on the help of Jenkins and her Lingua Franca Core (2000: 158). She proposes a number of pronunciation areas in which she believes errors should be eliminated so as to preserve intelligibility in international communication contexts. Using Jenkins’s findings and recommendations as a starting point, I developed a small-scale investigation into Brazilian students’ intelligibility (Silva, 1999), followed by an on-going process of data collection. The results of this investigation – somewhat surprising in many cases – can give us some idea of features of the way Brazilians pronounce English that are potential problems for intelligibility. The main ones, which cause serious misunderstandings seem to be:

  • 1. Confusion between the phonemes /r/ and /h/ in initial position:
    E.g.: *hõõ instead of room, *he-tir-ed instead of retired, *Hobbin instead of Robin
  • 2. Strong reduction of the vowel sound in final unstressed syllables (rendering it inaudible to non-Brazilian ears)
    E.g.: *coff instead of coffee, *offs instead of office, *happ instead of happy
  • 3. Stress placed on wrong syllables (especially in compound words) and on the wrong words in a sentence.
    E.g.: *classROOM instead of classroom, *homeWORK instead of homework, *like IT instead of like it.
  • 4. ‘Brazilianized’ pronunciation of loan words tranferred to English.
    E.g.: *how instead of hall, *cown-trie instead of country

Other pronunciation problems, such as the difficulty in pronouncing M and L in final positions – as in *(some) or *Bew (Bill) – do cause misunderstandings in only a few cases, and therefore seem to be less serious. Interestingly enough, the wrong pronunciation of TH sounds – many teacher’s ‘pet’ pronunciation problem – do not seem to cause any misunderstandings – *I sink (I think) and *sin (thin) are apparently no big deal as far as intelligibility is concerned! Neither is the adding of the extra /i/ sound after final consonants – *I likey (I like) and *wivvy (with). And this just confirms Jenkins’s Lingua Franca Core.

A lot more research and experimentation will be necessary until we can come up with a reliable list of what to focus on more intensely when teaching pronunciation so that our learners are really intelligible. This is a challenge we are faced with these days. Meanwhile, we must take what we already know into consideration, and try to answer our third question above. We may find that there is a lot we can already do to focus our teaching and help our learners give certain pronunciation areas – the ones that really matter – the importance they deserve.

Priorities in the classroom

I will list here a few ideas about how we can deal with prioritisation which may be useful when teaching pronunciation. These ideas are here more as food for thought and points to reflect upon than as ‘magical tricks’ to be followed, but teachers may decide to give them a try.

R versus H

  • Whenever you teach new words beginning with the letter R, make sure you remind students that they cannot pronounce them as in Portuguese. Make them aware of the importance of not pronouncing an R as an H by showing them minimal pairs, such as rat/hat, or red/head.
  • Constantly correct this mistake whenever it occurs. Make it into a ‘big deal’ in your classes. Don’t let students get away with H’s in place of R’s.
  • Have students produce posters to illustrate one or two minimal pairs: for example, a rat wearing a hat.

2 Many researchers, such as Shepherd (1987), have put forward comprehensive lists of common Brazilian pronunciation problems which are really helpful tools, though they do not consider any particular order of importance.

Reduction of the vowel sound in final unstressed syllables

  • Start by making students aware that in Portuguese we usually reduce the final syllables of words when they are not stressed. Give them a few examples – such as the word “caderno”, for instance. Make them say it naturally in Portuguese so that they see we hardly produce the final vowel, very often ending the word at the N sound – something like [ ] Say it yourself, making sure you pronounce it like that. Then tell students we cannot do the same with English words, as people from other parts of the world won’t be able to hear the ending of the words if we do that.
  • Show students minimal pairs like cough/coffee, or tax/taxi, and dare them to pronounce the words in them really different from one another. Make it into a game.
  • Be alert to this type of word when you teach new vocabulary. They are a lot more frequent than you think! Many English words follow a nnn or nn stress pattern, which means a large number of final unstressed syllables running the risk of being ‘swallowed up’ or ‘sucked in’ by your students.
  • Again, don’t let your students get away with doing that! Pretend you don’t hear the endings of words whenever they reduce them as they would in Portuguese, and repeat the ‘mutilated’ word with a confused, quizzical look on your face and a question intonation. Then pretend you ‘finally’ got what was meant, and repeat the word making the final syllable sound as it should. Then get them to repeat it correctly.

Wrong stress in words and sentences

  • You can never emphasize word stress too much when teaching vocabulary. Make word stress into another ‘big deal’ in your pronunciation teaching. Include the question Where’s the stress in this word? in your list of classroom language sentences, side by side with What’s ___ in English?
  • Constantly ask students to guess / predict the stress of new words.
  • Make sure you sensitise students to sentence stress as well. For example, don’t allow strong forms of object pronouns (e.g. I love HIM. / She went to the party with ME.) unless it is done for emphasis.
  • Do not take wrong sentence stress for granted! Be alert.

'Brazilianized' pronunciation of loan words

  • Start with your students a list of English words that are also used in Portuguese to be hung on the classroom wall. As you add new words to the list, raise students' awareness to the fact that the way we pronounce them in Portuguese is not necessarily the way they should be pronounced in English.
  • Create two different symbols to write next to the words as you put them on the list: one meaning “similar pronunciation in English and Portuguese” and another one meaning “different pronunciation in English and Portuguese”.

What about the TH sounds and the rest of the pronunciation problems? Should we just ignore them? No, I do not think we can afford to do that. It would not be fair on our students not to give them a chance to try all possible pronunciation features! Even if the correct pronunciation of the TH is not so important for intelligibility, it is our duty as teachers to teach. We can do it in a way that will allow those students who are able to pronounce it to get a chance to do it well. But there's no need to overdo it and spend precious classroom time trying to make everyone master it. Instead, we should teach those who are less “fortunate” – i.e., unable or unwilling to put the tip of their tongues “between their teeth”, to produce an acceptable alternative, such as T, for example (think of the Irish who get away with “I tink” and “someting” and “Tursday” from cradle to tomb).

In short, what really matters is what we teachers can do to help our learners become as intelligible as possible, and not so much what we should stop doing. There is no pronunciation feature we should not teach. But it has to be very clear to us that not all pronunciation features are equal, and that the ones that are made superior to others are the ones that really make a difference.


BRAZIL, D. 1994. Pronunciation for Advanced Learners of English – Teacher’s Book. Cambridge: CUP.

GNUTZMANN, C. 2000. ‘Lingua franca’. In: Byram, M. (ed.) The Routledge Encyclopedia of Language Teaching and Learning. London: Routledge. 356-359.

JENKINS, J. 2000. The phonology of English as an international language. Oxford: OUP.

SHEPHERD, D. 1987., ‘Portuguese Speakers’. In Swan & Smith, Learner English. Cambridge: CUP. 90-93.

SILVA, R. 1999. ‘A small-scale investigation into the intelligibility of the pronunciation of Brazilian intermediate students’. In Speak Out! 23: 19-25.

WALKER, R. 2002. ‘Choosing a Model for Pronunciation - Accent Not Accident’. TESOL Spain Newsletter No. 25: 8-9.

Ricardo Sili
Ricardo Sili has been a teacher and teacher trainer for many years, and is one of the authors of the Interlink Series. He is currently the editor of the Connexion Series.