The terms prescriptive and descriptive are well known with reference to the teaching of grammar. However, in this article they are used in regards to the teaching of pronunciation. Here, Adrian Tennant looks at a number of issues surrounding the ways in which pronunciation is dealt with in learning materials and by many teachers, as well as giving some practical suggestions on how to include descriptive pronunciation in your classroom.


One of my students recently asked me if they should say /ˈvɪtəmɪn/ or /ˈvaɪtəmɪn/. I told them they could say either and then asked them why they wanted to know. They told me that their teacher at school had insisted they should say /ˈvɪtəmɪn/, but that they’d heard lots of people saying /ˈvaɪtəmɪn/. This is one case where being prescriptive in terms of pronunciation seems utterly absurd. Here the difference is simply that one is British English pronunciation, whereas the other is standard American English. However, when it comes to choosing what pronunciation to teach, is it as simple as deciding whether to teach Received Pronunciation (RP), Standard American, or another ‘accepted’ native form?

For the moment, let’s put the socio-political issues to one side and focus on what we actually mean by teaching a standard form of pronunciation. To start off with, if we take RP, only about 3% of the population of the UK actually speak with such an accent (that’s around two million people) – not very many when you take into account the number of English speakers around the world. Therefore, should something like RP be taken as a model? Even if we take Standard American, which is spoken by more people than RP, can this also be used as a model?

Firstly, it is very hard to teach a particular pronunciation unless you yourself actually speak using that pronunciation. It may not be impossible, but unless you are using recordings as a model and not opening your own mouth, it would be extremely problematic – even if it is seen to be desirable (which in itself is very questionable).

Secondly, is a model such as RP really desirable? I’m often amazed when I watch international gatherings such as the EU or UN where English is frequently used as the language of communication. Quite often you’ll see a non-native speaker using English and most of the other participants listening intently, nodding, frowning, scribbling notes, etc. And yet, as soon as the British or American representatives start to speak those same people reach for their headphones in order to hear the translation! It may well be, that models such as RP and Standard American are actually far harder to understand than other forms and accents of English.

Thirdly, even if we could find an acceptable model for everyone, how accurately could we describe it? When it comes to what we teach at the moment, there are many cases where the pronunciation ‘rules’ or models, simply do not match what is actually used by most speakers. This is not always a matter of speakers having different accents (or not speaking RP or Standard American), but is often because the prescriptive rules are not a true reflection of use. Let’s take a look at some examples.

Does it sound like that?

It’s not just between American and British English that there are variations in the pronunciation of sounds within a word. In fact, there are many variations between different regions in Britain, let alone from country to country. Even people who seemingly speak the same variety of English will often have slight differences in the way they pronounce sounds.

When I look in the dictionary, I’m almost always given one phonemic transcription as though there is only one way to say a particular word. If I’m lucky, I’ll be given two – RP and Standard American. Not only is this a very false picture, but it is even a simplification within that particular model. Here are just a couple of examples: At the moment I live in Nottingham in central England. People here will talk about /græs/ (grass) whereas in Oxford they would talk about /grɑːs/, which is what my dictionary shows. In Nottingham, people take the /bʊs/ (bus) whereas the dictionary tells me it should be /bʌs/. We’re talking about a distance of around 175 kilometres between Nottingham and Oxford and yet they seem to speak a different language!

These differences don’t arise because the words are part of connected speech. They arise simply because people have a different accent, a different standard, a different model. It appears that the further north you travel in the UK, the shorter the vowel sounds get in words (although this is a broad generalization).

Jokes often rely on word play, homonyms, homophones, etc. Here’s one that uses pronunciation to create the joke:

A man walks into a bar with a giraffe and they both have a drink of beer. Afterwards, the giraffe falls over and goes to sleep. The man leaves the bar without the giraffe. The barman then runs after the man and says, “Hey! You can’t leave that lying there.” “It’s not a lion, it’s a giraffe,” says the man.

Now, how does this work? Well, if you say the sentences at normal speed then the word lying in the sentence the barman says comes out as /laɪn/ which sounds closer to /ˈlaɪən/ (lion) than /ˈlaɪɪŋ/ (lying).

Of course, what’s happening here is a feature of connected speech. But, this is precisely why prescriptive rules as to how words should be pronounced are pretty much unworkable. What we need to be doing is exposing our students to a variety of accents and ways of pronouncing words and making them aware that there are differences. 

Is it only sounds that are a problem?

No, not at all. Stress patterns, for both words and sentences, and intonation can cause serious problems. Again, the prescriptive rules are often wrong. Usually this is due to oversimplification. A good example of this is with intonation and questions. Reading many books, you would be under the impression that intonation always rose at the end of questions. Some books try to be a little more specific and state that for Yes/No questions the intonation rises, while for wh- questions it falls. However, even this is a simplification. In fact, questions are often marked simply by the use of specific ‘question’ words or by where the stress falls. Intonation (and to a certain extent, stress) is used to indicate other things such as mood, certainty, etc and not simply whether the sentence is an interrogative or a statement.

Why is this important?

The issue is not so much whether your students are going to be understood or not (although we can’t totally ignore this), but whether or not your students are going to understand other people when they are the listener. Simply listening to different accents is not enough – although it is certainly better than only hearing one accent, whether that is RP or something else. What students really need is focused work where the emphasis is on awareness and identifying differences in pronunciation. In other words, they need to be exposed to a descriptive approach (describing through action, in this case) to pronunciation rather than a prescriptive one (’this is how x is said’).

Some practical ideas

Making students aware that there is not necessarily only one way to pronounce things is important. In fact, in most languages there are variations, so it shouldn’t come as a surprise that there are many in English. Here are a few ideas in terms of raising awareness:

1. US or UK? Either record or write up the phonetic transcriptions of a number of words that have sounds or stress patterns that differ between American English and British English. Ask your students to listen to (or say) the words and note the differences.

2. What do you hear? Whenever you do a listening task, reuse the material focusing on the pronunciation aspects and not simply on the content. There are a number of different things you can do:

a) Ask students to listen to words that sound ‘unusual’. Focus on these words and try and work out together what is different. Is it a particular sound that is pronounced differently? Are there weak sounds, i.e. schwas? Is the stress on a different syllable?

b) Transcribe it. Ask students to listen to a short phrase or sentence and write what they hear. If they know how, you could get them to do it using phonetic script.

c) How else could it be said? Focus on some words or phrases that can have different pronunciation. Write down how it could be said and then try saying it using the alternative pronunciation.

3. I’m comfortable with … / I like her accent: Whenever students come across alternative forms of pronunciation get them to try saying it in the way they hear it. Ask them if they feel comfortable pronouncing the word or phrase in that way.

Note: Nowadays coursebooks tend to have a wide range of accents in the listening materials. Use this variety to help your students try out different accents rather than just be passively exposed to them.