Ken Paterson Logo

Let us teach (some) spoken grammar!

11 September 2019

The article below was published in The Teacher in September 2019

To find out more about my online course, 'Spoken Grammar: A Guide for English Language Teachers', please click here


Let us teach (some) spoken grammar!


What is spoken grammar?

By the early 1990s, corpus research – the computer-processing of millions of words of written and spoken English – had begun to discover a number of frequently-used items of conversational grammar that rarely featured in traditional teaching materials. Here are six examples, with the key language in bold (where appropriate), and a brief description in brackets:

1. What are they waiting for, those people over there? [a ‘tail’]

2. A: Lovely day, isn’t it? B: Beautiful. [initial ellipsis]

3. I saw her at Jo’s. That’s when she told me she was pregnant. [a demonstrative wh- cleft]

4. Come in. It’s lovely and warm inside. [binomial adjectives]

5. He’s a shortish kind of guy with browny-black hair. [vague category language]

6. So I said, look, it’s your decision in the end. [a spoken discourse marker]

Initial results of this research were written up by Michael McCarthy and Ronald Carter in a fascinating1995 article for the English Language Teaching Journal, titled ‘Spoken grammar: what is it and how can we teach it?’

So, in answer to the question at the top of this section, a definition of the term might be ‘lexico-grammatical elements of conversation that have been described by corpus researchers (and noticed by teachers), but haven’t yet become part of our everyday teaching syllabuses – or where they have, deserve greater prominence because of their frequency.’


Two key publications, and two questions

Among the commentary that followed this burst of ELT spoken corpus research, two major publications, which you may well be familiar with, emerged: the ‘Longman Grammar of Spoken and Written English’ by Biber et al in 1999, and the ‘Cambridge Grammar of English’ by Ronald Carter and Michael McCarthy in 2006. Both of these corpus-based grammars contain long  – and stimulating – passages on conversational grammar.

But these passages, interesting though they may be, also raised questions for the English language teaching community: is this the kind of language that we should be teaching? And if it is, can we agree on what to teach?


The pros and cons of teaching spoken grammar

On the ‘pro’ side, the very fact that these new items have risen to the surface in a process of frequency analysis indicates that they are likely to play a useful role in the interactive business of everyday conversation. So why shouldn’t our students have access to them? On the ‘con’ side: are we running the risk, in an era of global English, of imposing new native-speaker forms on the already-overburdened learners in our classrooms?

In the end, we probably need to make up our own minds, based on our instincts as teachers and materials writers, and the needs (and desires) of the particular students we’re teaching. But if we do want to go ahead, how do we decide, from the research available, which items of spoken grammar to teach?


In search of a syllabus

What we need, I think, as we sit down to study the published data, is a simple criterion that we can apply to any specific item that may interest us: would it have a useful function in the everyday conversations that our students take part in?

On applying this criterion, we can quite quickly rule out items which any good descriptive grammarian would need to include in a survey of spoken English, based on frequency of use, but have, we can probably all agree, little or no pedagogical value. Examples of these might include common ‘dysfluent’ speech acts such as ‘clausal blends’, where the speaker, under real-time pressure, mixes two clause types, e.g. ‘In fact, that's why last year they rented a nice house, in er Spain, was it, is that it was near the airport.' (from ‘Cambridge Grammar of English’, p.171). Excluded, too, might be socially or regionally marked forms like the common invariant question tag ‘innit’.  

But after that, we enter a greyer, more subjective area, where we need to make our own decisions. I’ll show you what I, as a teacher and materials writer, came up with.


A sample syllabus

In my recent online course ‘Spoken Grammar: a Guide for English Language Teachers’ (Udemy 2018), I finally settled on 15 items or features, organised under five headings:


1. Heads and tails (relocating language elements to the beginning or end of an utterance), e.g.:

The hotel where you’re staying, is it in the centre of town?

They’re a bit tight, these shoes.      

2. Declarative questions (statements used as questions), e.g.:          

So I hear you’re getting married?

3. Ellipsis, e.g.:                                                                      

A: You going to the match on Saturday? B: Don’t know. Depends on my work schedule.


4. Hyperbole (exaggerated language), e.g.:   

I’ve got a million emails to answer.

5. Interjections (‘Oh’, ‘Wow’, ‘Ouch’ etc.), e.g.:

Oh, before I forget, have you got Sara’s number?

6. Cleft structures and binomial pairs, e.g.:

Why don’t you try Luigi’s? That’s where you get the best sandwiches.

How was the meeting? Nice and quick, I hope?


7. Vague categories, e.g.:

What’s happening over there? Has someone won the lottery or something?

8. Vague placeholders and quantities, e.g.:   

Have you got that thing for cutting pizzas?

I just need a bit of peace and quiet.

9. Vague lexical bundles (very frequent word ‘strings’ with an uncertain intention), e.g.:

I don’t know why you said that.


10. Adverbials (anyway, then, though, actually, really, of course, just), e.g.:       

You’ve been friends for a long time, then?

I’ve got a bit of bad news, actually.

11. Discourse markers (I mean, you see, you know, right, well, listen, look, hey), e.g.:

Listen, why don’t we just accept we’ll never agree on this?

12. Ways of using direct speech (in conversation), e.g.:

So she insisted on paying for the meal and I said, okay, as long as I get the next one.


13. Response tokens and response questions, e.g.:

A: … and finally, you turn left next to a French restaurant. B: Great. Fantastic. Thanks very much.

A: I resigned today. B: You did what? A: I just couldn’t stand it any longer.

14. So and do (special uses), e.g.:

A: I’m selling my flat. B: So you told me. Have you put it on the market yet?

A: I’m going to order a taxi. B: No, don’t do that. I’ll give you a lift.

15. Synonymous language (immediately rephrasing an element of something you’ve just heard) and dependent clauses (‘completing’ what you’ve heard with an appropriate clause), e.g.:

A: It’s very hard to find a part-time job these days. B: I know. It’s a real struggle.

A: We’re going to miss the start of the film. B: Unless we get a cab.



As with any conventional feature of grammar, spoken grammar items have rules or patterns of usage that learners need to know. One of the challenges, however, in this new and evolving field, is that, although some published guidance is available, we may, from time to time, have to work things out for ourselves. In that spirit, try this short quiz! (Suggested answers at the end).

What is the function that the words in bold in each of these pairs of examples have in common?

1. [Heads]

That white building, is it the Museum of Contemporary Art?

The guy you spoke to, did he give you any advice?

2. [Binomial pairs]

Let’s hope the journey’s nice and easy.

Use this rope. It’s good and strong.

3. [Ways of using direct speech]

He looked a bit down, so I said, hey, are you OK?

I told her it was my birthday and she said, oh, I’m sorry, I completely forgot.


Teaching and practice

Once you accept the ‘legitimacy’ of spoken grammar, there’s no reason why it shouldn’t be taught and practised like any other language item. One thing that I found particularly useful in creating materials for my online course were ‘compromise dialogues’: short or longer made-up (but naturalistic) conversations that are deliberately ‘loaded’ with the target items.

Here are examples from a section focusing on three vague category phrases:

… and everything, … or something, … and stuff

A: Is Katie really going into business, then?

B: Of course. She’s bought a shop, hired staff and everything!

A: Amazing! But it’s fairly risky, isn’t it, from a financial point of view?


C: We’ve got about an hour before we need to leave for the airport.

D: OK. Shall we go for a breath of fresh air or something?

C: Sure. I’ll get my coat.


E: Did you buy anything at the market?

F: Yeah. I got some fish, potatoes and stuff.

E: Great. Get cooking, then!


Dialogues like this can be used in many ways, including contrastive analysis; commenting or discovery activities; drilling and controlled practice (with gap-fills etc.); and may even lead on to short role-plays and simulations.

One thing I think you’ll find, if you do introduce your students to a range of spoken grammar items, is that learners actually enjoy using them, and will tend, later on, to bring them into their regular conversations.


Suggested answers to the quiz:

In 1., ‘it’ and ‘he’ are ‘co-referential’ pronouns, representing the ‘heads’ (‘That white building’ and ‘The guy you spoke to’) in the second part of the question. In conventional word order, these pronouns would not be necessary: ‘Is that white building the Museum of Contemporary Art?’, ‘Did the guy you spoke to give you any advice?’.

In 2., the adjectives ‘nice’ and ‘good’ don’t really add anything in terms of meaning. Instead, they have a positive, gently emphatic effect in combination with ‘easy’ and ‘strong’. (Verb binomials, where the second verb in the pair carries the essential meaning, are similar, e.g. ‘Come and eat something.’)

In 3., both speakers incorporate direct speech into their utterances, rather than reporting it with tense and pronoun shifts. Biber at al (1999) suggest that marker words such as ‘hey’ and ‘oh’ (whether or not they were actually used by the original speakers) may well be used like speech marks to indicate the beginning of a conversational ‘quote’.



1. Publications in the order mentioned above:

McCarthy, M. and R. Carter. 1995. ‘Spoken Grammar: what is it and how can we teach it?,’ ELT Journal 49(3): 207-218.

Biber, D., S. Johansson, G. Leech, S. Conrad  and E. Finegan. 1999. The Longman Grammar of Spoken and Written English. Longman.

Carter, R. and M. McCarthy. 2006. Cambridge Grammar of English. Cambridge University Press.

Paterson, K. 2018. Spoken Grammar: a Guide for English Language Teachers.

2. Further reading:

Biber, D. 2009. ‘A corpus-driven approach to formulaic language: multi-word patterns in speech and writing,’ International Journal of Corpus Linguistics 14: 381-417

Carter, R. and M. J. McCarthy. 1995. ‘Grammar and the Spoken Language,’ Applied Linguistics 16(2): 141-158.

Carter, R. and M. McCarthy. 2004. ‘Hyperbole in everyday conversation,’ Journal of Pragmatics 36: 149-184

Carter, R. and M. J. McCarthy. 2015. ‘Spoken Grammar: Where Are We and Where Are We Going?,’ Applied Linguistics 2015: 1-21

Carter, R., M. J. McCarthy, G. Mark, and A. O’Keeffe. 2011. English Grammar Today. Cambridge University Press.

Channell, J. 1994. Vague Language. Oxford University Press.

Goh, C. 2009. ‘Perspectives on spoken grammar,’ ELT Journal 63/4: 303-12

Norrick, N. R. 2012. ‘Listening practices in English conversation: the responses responses elicit,’ Journal of Pragmatics 44/5:566-76

Paterson, K., C. Caygill and R. Sewell. 2011. A Handbook of Spoken Grammar. Delta Publishing.

Timmis, I. 2012. ‘Spoken language research and ELT: where are we now?,’ ELT Journal 66/4: 514-22