G is for Grammar McNuggets

18 09 2010

Photo: Frtiz Saalfeld

Stephen Krashen once said (only half-jokingly, I suspect) that, more important than having new ideas is giving old ideas new names. With that in mind, I was reminded recently that it was 10 years ago that I coined the term “grammar McNuggets” (in a talk at IATEFL Dublin in 2000). Essentially, there is nothing new in the view that grammar is artificially packaged into bite-sized chunks for the purposes of teaching: William Rutherford had used the term “accumulated entities” in a book in 1987, and who knows how long the term “discrete items” has been around? So, why “grammar McNuggets”?

What I wanted to capture was not just the discrete-item nature of the grammar syllabus, but the way that this is exploited, particularly by publishers, for the purposes of the global marketing of EFL. To do this, I drew on a construct, familiar to students of cultural studies, and first developed by Stuart Hall, called “the circuit of culture”. The circuit of culture is a construct for the analysis of cultural artefacts that has been applied to a range of objects, including the Sony Walkman. Du Gay (1997), for example, argues that

to study the Walkman culturally one should at least explore how it is represented, what social identities are associated with it, how it is produced and consumed, and what mechanisms regulate its distribution and use. (p. 3)

Applying this model to pedagogical grammar, I was curious to see how grammar is represented (e.g. in publishers’ catalogues), how it is produced — or better — reproduced, how it is consumed in the classroom, how it is regulated (e.g. by exam boards), and who identifies with it (e.g. what ideas and values are associated with an allegiance to grammar teaching).

With regard to its (re-)production, I was drawn to this text on ‘McDonaldization’:

A perfect example of a simulated product is McDonald’s Chicken McNugget. The executives at McDonald’s have determined that the authentic chicken, with its skin, gristle and bones, is simply not the kind of product that McDonald’s ought to be selling; hence the creation of the Chicken McNugget which can be seen as inauthentic, as a simulacrum. There is no “real” or even “original” Chicken McNugget; they are, and can only be, simulacra. (p. 10)

To quote from the text of my talk: “Much of what is taught as pedagogic grammar is of equally doubtful authenticity. The skin, gristle and bones of language have been removed such that “grammar exists independently of other aspects of language such as vocabulary and phonology” (Kerr, 1996: 95). Moreover, the findings of corpus linguistics in particular suggest that pedagogic grammars only loosely reflect authentic language use and that “some relatively common linguistic constructions are overlooked, while some relatively rare constructions receive considerable attention” (Biber, et al. 1994, p. 171). An enthusiasm for compartmentalization, inherited from grammars of classical languages, has given rise to the elaborate architecture of the so-called tense system – including such grammar McNuggets as the future-in-the-past, and the past perfect continuous, not to mention the conditionals, first, second and third – features of the language that have little or no linguistic, let alone psychological, reality. While attempts have been made to restore authenticity to grammar, such attempts have generally fallen on deaf ears. If some more recent coursebooks are anything to go by, grammar syllabuses are becoming less innovative and even more derivative”.

That was ten years ago. Is it still true?


Biber, D., S. Conrad, and R. Reppen, 1994. Corpus-based approaches to issues in applied linguistics. Applied Linguistics 15/2, 169-89.

du Gay, P., Hall, S., Janes, L., MacKay, H. and Negus, K. (1997). Doing cultural studies: The story of the Sony Walkman. Milton Keynes: The Open University.

Kerr, P. 1993 `The role of language analysis on CTEFLA courses’ in Future Directions in Teacher Training: Conference Report International House, London.

Ritzer, G. (1998). The McDonaldization Thesis: Explorations and extensions. London: Sage Publications.

Rutherford, W.E. 1987 Second Language Grammar: Learning and Teaching. Harlow: Longman.



48 responses

19 09 2010
Alex Case

I would say that the problem with some more recent coursebooks is that they have started to take some of that spoken grammar, lexicalised grammar etc and just add it on top of the grammar that we always dealt with like tenses and uncountables. With the limited time that leaves for each bit we get little more than one set of written exercises for each one. If that didn’t work when we were only trying to teach them a limited number of things, why would it suddenly produce results when they have five things per unit to cope with, some of which their previous teachers had never mentioned??

At the same time, publishers also seem to be producing whole courses based on the model that was formerly only used for short courses, being a page or two on one grammar point, right that’s done, turn to the next two page spread on something else. Again, what they cover is slowly changing, but if they teach each thing that badly, what do we get out of that?

20 09 2010
Scott Thornbury

Thanks Alex – for being the first cab off the rank!

I think you’re right – there’s a danger that the discoveries being made by corpus researchers will simply be ‘processed’ into yet more ‘grammar McNuggets’ for consumption in coursebooks. (I have an image of a vast underground mine, the corpus linguists beetling around down below with their little picks and shovels, while a continuous conveyor belt carries their findings up to the surface, where they are borne away by lorries labelled ‘Longman’, ‘Oxford’ etc).

On the plus side, corpus linguistics is now able to tell us which of the traditional grammar McNuggets are high frequency (and in which registers) and which are not. One recently published coursebook series adduces this evidence to justify the fact that the present simple is presented (in Unit 1) before the present continuous (in Unit 2) – although if it really took frequency findings literally, the present continuous shouldn’t really occur in the sequence until Unit 24!

19 09 2010
Alex Case

PS If I’m the first person to comment on a Scott Thornbury post, Twitter and RSS feeder free caveman that I am, maybe the blogging boom really is over!

19 09 2010

Well, I am just about to start using commercially produced textbooks after many years. What I notice is a strong preference on the part of authors for discovery techniques when it comes to grammar. So, instead of being told by the teacher something like ‘the Present Perfect is used for actions/states that began in the past and have continued until the present’, learners discover that fact. The question is ‘So what?’

What is that rule doing for the learners? What are learners supposed to do with it? Learn it? They are certainly supposed to apply it in the usual gap fills and eventual (often tediously artificial) production activity. (By the way, can you imagine how boring this endlessly repeated procedure must get for some students… year after year?) Somewhere along the line, with some revision, it’s all supposed to become automatic: out pops the present perfect unbidden from the mouths of our students. … Things haven’t changed that much.

Yet I live and teach in a city where students spend year after year after year consciously applying those rules and normally can do precious little with whatever English (or French) they have learnt at the end of it all; a city in which learning languages at school for 12 years is by and large a total waste of time. (Interestingly, the politicians governing the area have now responded to that situation by tacitly owning up to the fact that traditional language teaching just does not work here. How? By initiating a massive programme of bi-lingual education in which kids do virtually all their subjects through English from the word go.)

20 09 2010
Scott Thornbury

Thanks Glennie. Your doubts echo those of a teacher who was a respondant in a study that Simon Borg made of teacher’s attitudes and beliefs about grammar teaching:

“I’m trying not to exclude the possibility, perhaps the probability, that formal language focus at some point gets transferred into language which is acquired by the student…. [But] I don’t necessarily believe that it’s going to help them. I’ve done this present perfect umpteen times with a million people. I still believe that nothing I’ve ever done in a classroom consciously with students, language focus, has helped them to acquire the present perfect, for example.”

Borg, S. (1998). Teachers’ pedagogical systems and grammar teaching: A qualitative study. TESOL Quarterly, 32/1, 9- 38

19 09 2010
Marisa Constantinides

I like the term “McNuggets” – it describes the bittiness and artificiality of what passes for (grammar) teaching so well…

I have observed a fair share of lessons taught in this way, like anyone who is involved in initial level traininng (and higher, I should add) and have not seen the learners progress towards easier, more fluent speech in any leaps and bounds – quite the contrary…

For me, it is not just the pre-fab, pre-processed nature of these bite-sized chunks which is to blame for the learners’ lack of ability to progress and reach some semblance of fluency. The linearity of coursebook syllabuses is also an issue.

There is a ‘been there, done that” attitude in materials which fragments rather than defragments, scatters knowledge all over the place making it difficult to bind it together and make sense of it or make it accessible.

This points back to the materials writer, their editor as well as the marketeer behind them and I think you are right – they must have perceived a market trend, otherwise why does this mcnuggetism persist?

Newer coursebooks, as Alex suggests, adding the so-called ‘everyday language’ bit, paying lip-service to functional language are just merely selling more mcnuggets in the same small basket.

I am not against coursebooks, I hasten to declare, but I see more and more people either abandon them, or just use the texts and skip the rest.

Which, I think, brings Mr. Stephen Krashen back into the picture.


19 09 2010
Alex Case

Of course, grammar translation is probably used even more than textbook style grammar mcnuggets. And then there is the “take in a newspaper article and talk about it” that seems to be the main actual (rather than theoretical ) reaction against both of them. I find the last of those to be just as unmiraculous as any of the other common ways of teaching.

I personally have found a few grammar mcnuggets useful during language learning in my time, but I can only keep between one and three in my head at any one time. A few more pop back into my head years later when I come across the form again. The rest are lost, forgotten, or even confuse me more than being told nothing.

Whatever language it is you choose to deal with in textbooks, I still say that this is how it should be dealt with:


The main thing is that you should come across the form, have a little space to absorb it, look at it in more detail and start to produce it, have another little pause, and then expand on that knowledge and try again. With some other structures, you can extend the pauses, miss out some stages, etc, therefore having something that is still a predictable structure, but one that is flexible enough to deal with all kinds of language

24 09 2010
Scott Thornbury

Alex, sorry not to get back sooner – I like your idea of a more staged approach to grammar teaching (as outlined in your article), with an initial ‘seeding’ (or ‘priming’) stage, followed by successive ‘forays’ into closer analysis after a period of gestation. The ‘breadmaking’ metaphor is perhaps even more suggestive than the seed-planting one: the ingredients are mixed together, and let to sit (in a warm place) for a while, then they are kneaded, left to sit again, kneaded again, and so on, until the whole lot is ready to bake. Different ‘loaves’ will sit longer than others – depending on the rate of levening (is that a word?) and the exact consistency and texture desired or achievable. (But what – in classroom terms – is putting the bread in the oven, I wonder?)

24 09 2010

Getting the students to actually USE their loaves, I guess. Or their noodles.

19 09 2010

I do occasionally worry that my own English is gradually being compromised by the sometimes stilted grammar of the textbooks, and that it is supplanting my authentic English. However, this is probably a good thing for my students as this is the English they speak, and the English they encounter when they speak to other people using it as a foreign language. In that sense it is probably more authentic than the rather parochial stuff I grew up speaking.
Very interesting article Alex. I would be very interested to use such a textbook – more like a hyper-textbook by the sounds of it – are you going to write it?

20 09 2010
Scott Thornbury

Thanks Tony. It might be interesting to study the phenomenon of ‘teacher-ese’ – the characteristic speaking style that EFL teachers develop as a result of a lifetime spent teaching grammar McNuggets!

20 09 2010

Thanks Tony. It actually started as a proposal to a publisher, and I turned it into an article when they rejected it so all that effort didn’t go to waste. The feedback was something like “That’s not really what we’re looking for with this project, but we hope that doesn’t make you think that we are anti-innovation”.

It would be bloody hard work to design a multi-layered syllabus like that, so I’m hoping someone else will be inspired by the article to try it, therefore saving me the work

20 09 2010

It would be hard work if you did it by yourself, but it strikes me that Scott is setting something of an example here with his A-Z in terms of online collaboration. You could devise a framework and every week or so write up an idea and invite a response, rather like a textbook wiki. Wikis never really end, but you could have the makings of a course within a year or so, and one devised by some of the finest brains in ELT.

19 09 2010
Adrian Underhill

Thanks folks for the post and comments. I remember the delight with which I discovered and rapidly consumed an early edition of Thomson and Martinet during my first teacher training course, and my lack of any doubt that this manual described grammar at least as accurately as my VW manual told me everything about the workings of my Beetle. But like the people you refer to I see it differently now, and I wonder if McDonaldization is a one way street. Two points in response to your posting:

1. It seems to me that the grammar we teach is way downstream from the phenomenon it purports to describe, it takes place long after and far removed from the original event (one is systemic, one reductionist, one complete, the other incomplete, they probably even use different parts of the brain… I’m just guessing). Furthermore it characteristically describes the easily spotted, coagulated ‘planetary’ matter like conditionals and tenses, leaving huge stretches of dark matter between….

As Scott implies (“……grammar is artificially packaged into bite-sized chunks for the purposes of teaching …”) a critical look at pedagogic grammar could unpack a host of power structures and vested interests keeping it all in place and telling us as much about ourselves, its constructors, as about grammar.

2. In my case the need for a holistic approach to grammar teaching and learning has been much inspired by Gattegno’s approach, which is to let go of the traditional grammatical categories, start (nearly) anywhere with what engages the student and from there track and facilitate learning wherever it may lead, while exercising subtle guidance to keep the growth bushy. This of course does not work with a presentation-practice model, and requires a quite different set of teaching / learning techniques to service it. In fact it works rather badly with off-the-shelf direct method or communicative methodology. My point is that content and teaching method exert a pull on each other, with grammar McNuggets being matched by activity McNuggets

Like many I am attracted by corpus linguistics as maybe supplying ambient language in sufficiently concentrated form to stimulate learners’ intuitive rule-ready faculties to resonate with (complex) rules without trying to skewer them. But my personal answer to your question is yes, it is still true, and the next phase will be the McDonaldization of corpus linguistics.

20 09 2010
Scott Thornbury

“The grammar we teach is way downstream from the phenomenon it purports to describe…”

That’s a great image, Adrian, which I will immediately appropriate! In fact, it’s not dissimilar to the claim made by emergentists, to the effect that grammar is a kind of precipitate that, over time, ‘falls to the bottom’ and solidifies. Or, as Hopper puts it:

“Grammar is seen as … the set of sedimented conventions that have been routinized out of the more frequently occurring ways of saying things…”

Hopper, P.J. 1998. Emergent language. In Tomasello, M. (ed.) The New Psychology of Language: Cognitive and Functional Approaches to Language Structure. Mahwah, NJ.: Lawrence Erlbaum. 155-175.

20 09 2010
Nick Jaworski

I remembering first seeing the term and thinking it was relatively new about 7 months ago. Just goes to show how much the online community can teach you 🙂

I do love the analogy of the McDonaldization of English teaching (although the Baudrillard-esque quote on simulcra borders on postmodern absurdity I think). Language is reprocessed by schools, publishers, and teachers, packaged into bite-sized bits, and then sold for public consumption. I think we often like to focus on ELT, but this is quite a common tradition within the profession of teaching for hundreds of years and certainly existed before big corporations. Just think of your History and Math classes. I think there are as much major historical currents here of ideas in teaching as anything else. Bite-sized grammar was certainly around before McDonald’s. Perhaps we should be referring to the grammatization of food rather than the McDonaldization of grammar 🙂

20 09 2010
Scott Thornbury

Thanks Nick (nice to hear from you again!)

Yes, you’re right – MacDonaldization is not confined to the ELT domain alone, and in fact charcaterises an approach to education that is consistent with what Henri Giroux calls the “culture of positivism” (Giroux, 1997) in which “knowledge becomes identified with scientific methodology and its orientation towards self-subsistent facts whose law-like connections can be grasped descriptively”. Facts, after all, are easy to teach, and even easier to to test.

It’s the same educational philosophy that Paulo Freire called ‘the banking model’ of education, where the teacher ‘deposits’ packets of knowledge in the learner’s account.

20 09 2010

I think grammar Mc Nuggets are as prevalent today as they were 10 years ago. Many initial teacher training courses still spend a lot of time looking at the Mc Nuggets and the PPP lesson shape is still very much alive and kicking it seems. (or at least from what I’ve observed)

I hear it all the time that students “want more grammar”. And teachers then turn to the past perfect continuous and “give more grammar” yet the students are still unhappy – “but we still need more grammar” they say.

From what I’ve experienced what students tend to mean when they say “we need more grammar” is “we need the language to better enable us to perform this task or communicate in this situation” Which is where a TTT shape of lesson in my eyes is just much more effective. The language they need arises from the task and may or may not be a Mc Nugget (more often it isn’t) here we can bring in emergent language and lexical chunks too.

It’s so painful to see a class having to “produce” in a contrived way the 2nd conditional in a speaking activity. “So what would you do if you won a million pounds?” “Well, if I won a million pounds, I’d buy a yacht and what would you do if you won a million pounds?” I mean who really talks like that?

20 09 2010

After 22 years in the ‘profession’, I talk like that. My family are sick and tired of me looking out of the window and saying ‘Look at those clouds. It’s going to rain.’ And every time I have a shower, the phone rings! 🙂

21 09 2010

lol Glennie – must admit, I sometimes catch myself and then have to relax and speak “normally” – that’s a good one, grammar Mcnuggets are slowly destroying EFL teachers natural spoken discourse!

21 09 2010
Simon Greenall

Hi Scott,

Nothing to disagree with … except that it’s not just the publishers who promote grammar McNuggets, but Ministry of Education people around the world, and teachers who, through no fault of their own, don’t know how to teach the holistic and/or corpus based approach to linguistics. Publishers follow the market, they don’t create a market where none exists, and I suppose I’m a bit bored with carrying the can for promoting grammar McNuggets throughout my writing career.
Every time you challenge a teacher’s knoweldge of grammar, bite-sized chunks or not, you challenge their competence. There are millions of teachers who are not aware of this very well-argued criticism of grammar McNuggets, and who, I suspect, may not understand how it could be applied to their classes of 50+ students.
Your excellent arguments are, for me, always diluted when you take a swing at the publishers and coursebook writers.
It’s not just the publishers who are at fault with promoting grammar McNuggets, it’s the MoE and the trainers’ responsibility to get the message across.
It’s not the message that’s wrong, it’s the medium.


21 09 2010
Scott Thornbury

Yes, a fair point (as always), Simon. I think that the analogy with junk food consumption can be extended to support your line of thought: just as it’s not entirely the parents’ fault if a child is addicted to chicken McNuggets (it’s the whole food and leisure industry that is implicated), so too it’s not the coursebook writers’ fault that their books are part of a conveyor-belt educational culture where knowledge is packaged and processed. I know this only too well, having come up again and again (in my coursebook writing days) with demands from publishers – which were simply echoing demands from the market – that there ‘wasn’t enough grammar’ – or, worse, that the grammar that there was was in the wrong order, or even on the wrong page! (My petulant reistance to these demands sealed my fate as a coursebook writer at a relatively early stage!)

22 09 2010

This is a really interesting discussion for teachers and materials developers alike. “McNuggets” have two components: the ‘Mc’ implying they are over-processed, artificial, and a long way from the original chicken; the “Nuggets” implying “bite-sized and easily consumed” … The first is undoubtedly a problem, but is there such a problem with the second? As a language learner I always found that when some highly concentrated bite-sized bit of language ‘clicked’ it can serve to unlock a bigger system. (But only if I could recognise it as taken from real life & close to something I might need to use in the future). Maybe corpus-based material, if it is also alert to the learners’ most likely needs, can lead to the preparation of more nutritious, raw, bite-sized portions: “ethically-sourced grammar sushi” perhaps …?

22 09 2010
Scott Thornbury

Thanks for the comment, Keith. “Grammar sushi” certainly has a nicer ring to it than “grammar McNuggets”! Though I suspect that grammar is more like ramen than sushi – i.e. lots of intertwined stuff in a nutritious soup!

Yes, I’m not denying the generative power of some nuggets of grammar, and I often cite Schmidt and Frota’s study when, as a learner of Brazilian Portuguese, Schmidt found that “pre-digesting” some grammar McNuggets ‘primed’ him for subsequent noticing of these items in the input, when he actually hit the streets and started interacting in earnest. More work needs to be done, though, on researching which (Mc)nuggets really do have a beneficial effect, and which simply clog up the system.

23 09 2010

Thanks very much for your reply Scott. Grammar noodles is a great metaphor. What’s the title of the Schmidt and Frota study? It seems to describe my own language learning experience precisely.

Now, is there a food metaphor for dogme? Foraging for wild mushrooms with an expert guide, maybe …?

23 09 2010
Scott Thornbury

Hi Keith –

The reference is:

Schmidt, R., and Frota, S. (1986). Developing basic conversational ability in a second language: A case study of the adult learner of Portuguese. In Day, R. (ed.), Talking to learn: Conversation in Second Language Acquisition. Rowley, MA: Newbury House.

22 09 2010
Luiz Otavio

I think we’re basically talking about two different things in this thread: content and process. Yes, I do worry about the McNuggetation of grammar (Scott, brilliant term!) and the marginal, end-of-lesson role that functional language has played in recent coursebooks (and had it not been for the recently rekindled interest in the “Lexical Approach” it might not even be included at all), but I tend to be more critical of HOW coursebooks over the last few years or so tend to approach grammar: blind adherence to “discovery” approaches, a clear preponderance of gap-fill type activities, more emphasis on language manipulation rather than language creation, very little recycling. It seems ELT is much more linear and atomistic than it was, say, twenty years ago.
But that doesn’t necessarily depend, I think, on how textbooks package the grammar they want to teach. In other words, a lesson in which “might have” is taught, say, to enable students to admit guilt (“I might have said more than I should. Sorry”) rather than talk about why dinosaurs might have gone extinct may or may not suffer from the same problems I described above.
Just to draw a (somewhat far-fetched) analogy: didn’t someone say, back in the early 80s, that a functional syllabus is not necessarily tantamount to a communicative approach to language teaching? In other words, functions (or grammar!) may or may not be taught “communicatively”. Was it Keith Johnson?
By the way, is the Keith in this thread Keith Johnson? He was my MA tutor in Lancaster in 1999!

22 09 2010
Scott Thornbury

Hi Luiz, and thanks again for your thoughtful comment. Can I clarify? I’m not sure how to take your ‘might have’ example: are you saying it would be better to contextualise this in terms of self-exculpation (“I might have done X, if so I’m sorry”) or in terms of past speculation (“Dinosaurs might have been wiped out by a meteor”). Either way, it seems to me (but I would need to check a corpus) both contexts seem a little rarefied, compared, say to “Where are the keys?” “She might have left them under the mat”.

“It seems ELT is much more linear and atomistic than it was, say, twenty years ago”. This is also worth following up… in another comment!

22 09 2010

Hi Luiz – I’m not Keith Johnson, though I’d agree with the remark that you quote!

22 09 2010

Nuggets are easier to handle (and digest) than real live chickens running around the place. My favourite course? Headlessway.

22 09 2010
Simon Greenall

This is an incredibly important posting, and I’ve found everyone’s comments really interesting. The sensitive issue is that teachers wear their knowledge of grammar as a badge of their competence, and even today, we challenge this at our own risk.

So, it’s all fine and dandy to question the atomistic and linear qualities of ELT and the textbook course grammar syllabus. I’d LOVE to find a viable alternative.
But putting aside the practical issues of training teachers to eschew the Mcnuggets/sushi approach to grammar, and of classroom management where class sizes reach 50+, which poses problems for individualization, and the fact that classroom teaching in its own right is chronological and linear … it’s not just a textbook syllabus problem … what would be the overarching system, the fundamental, holistic principles of an alternative approach, or ‘diet.’? Or have I missed something.

It really interests me, because I could … I really could … stick my neck and put this all in a textbook. But I don’t know what the alternative is that you’re advocating. If it’s input + 1, well, fine for the second language context, but probably not for the foreign language (as was) context.

Only today I was asked to advise on an adaptation of a university level course, which at present has a ‘syllabus’ of what we’ve critically described as ‘peripheral grammar’, barely stable or generative structures which are on the upper edge between grammar and lexis. I don’t believe in it, but I now have the chance to change it. Now, the feedback suggests that for the new regions (Middle East and East Asia) teachers want a skills based course, but I’ve done this so many times elsewhere, and teachers still ask for more (traditional grammar.

So what is it you and everyone else, are suggesting I could do? I’ll cut you in on a share if you come up with something practical 🙂


26 09 2010
English Raven

Hi Simon,

In all honesty, I think this is why my own coursebook series has done so well. First in East Asia, then throughout Asia, then pretty much globally. It’s a skills series where Reading, Writing, Speaking, Listening, Grammar, and Vocabulary have been divided into separate strands. Each strand does integrate other skills, but has a general macro skill as its focus.

In this approach, I’ve gotten away with not having a grammar syllabus, because the only strand that features a grammar syllabus is the grammar strand (funny, that…). There is “topic streaming” across the strands, but basically, for the teachers/classes who want their Grammar McNuggets, there is a special strand of separate books just for that. The other strands focus on skills, with grammar being incidental or only drawn on where it highlights or enhances particular skills.

The books are selling – especially speaking and writing strands, and they don’t have any sort of specific grammar syllabus. Whether that is a reaction to having had too much grammar in the past or an indication ELT markets ARE willing to use other approaches that are not grammar-centric, I can’t really say at this point.

I mentioned this at length in a post from a while ago called “Exceptions to Thornbury’s Rule?”, which in turn was a reaction to some of Scott’s comments in the IATEFL YL-SIG discussion about coursebooks (which I think you (Simon) may have moderated?).


Anyway, I think this is a potential way forward. Put grammar in its own takeaway bag and sell courses by skill strands. Let teachers and schools choose as they like, but don’t force the grammar McNuggets into skill strands that don’t need them to be predominant, and let the more manufacturing types use it as a stand alone feature (which can still be topically integrated with the other strands).


– Jason

26 09 2010
Scott Thornbury

Thanks, Jason. The idea of ‘separating out’ the grammar, so that it co-exists as one of several independent strands in a course is an interesting one, and it’s encouraging that this approach has been well received. But isn’t there a slight danger of throwing out the proverbial baby? That is to say, if the skills strands have no overt (either prospective or retrospective) grammar focus, where is the ‘focus on form’ that is arguably necessary for language development? I’m not advocating no grammar, so much as no grammar-led syllabus, especially one that consists entirely of the canonical McNuggets.

26 09 2010
English Raven

Oh, there is certainly grammar in the other strands — of course, because there is language in them 🙂 — and in places certain kinds of grammar are highlighted as part of noticing procedures (where they fit or demonstrate the skills), but I daresay a specific focus on form would have to come through the approach and handling of the teacher. So sorry, when I said there is no overt *grammar syllabus* in the other skill strands, this was not to say there is not any real grammar in them.


22 09 2010

Hi Scott,
Yes, I was just trying to think of a more plausible context and the dinosaurs example was what occurred to me (as a classic example of contrivedmess/ mcnuggetization) when I was writing.
“She might have left the keys at home” sounds perfectly natrual, of course, but more and more I’ve been stumbling upon examples of self-exculpation. I was looking at the scripts for seasons 9 and 10 of Friends the other day and I ran a search for “may” and “might”. A few surprising discoveries: 1- In some of the episodes, there’s not a single occurrence of might (present or past) 2- “may have” and “might have” are used slightly more often for self-exculpation rather than speculation about the past.
I don’t want to generalize beyond those 30 something episodes, of course, nor would I regard them as representative samples (since there might be such a thing as genre-bias?). It’s just that this use of “modal perfects” was at the back of my head as I was commenting on McNuggetization.
I don’t think I’ve ever told you this, but I absolutely adore this blog.

23 09 2010
Scott Thornbury

(Thanks for the feedback, Luiz – but a blog is only as good as the comments it gets!)

23 09 2010

You’re right. Coursebooks can only go so far. Even if we were to write a series for experienced teachers teaching groups of 10 or fewer, there are certain things that are inherently incompatible, I think, with a book.

I think most (attempts at) task-based textbooks are good examples of how hard it is for a mainstream book to accommodate the demands of an organic, student-generated syllbus. The Willis’ Cobuild Series, David Nunan’s Atlas and, more recently, Cutting Edge deserve a lot of credit for trying to be task-based, but honestly, are they really? I don’t think so, nor do I think they could ever be.

The point I made earlier about ELT being more linear and atomistic refers more to classroom processes than syllabuses, I think. Recent coursebooks seem to believe that students move from declarative knowledge to procedural knowledge (Scott, I know you’ll hate me forever for using this terminology in your blog…) by being asked to notice a discrete language item in a text, answering a few discovery questions, doing a couple of gap-fills and then (if you’re lucky) a less controlled activity.

Coursebooks will always be at odds, I think, with the sheer “organicness” of language learning, but do they have to move so radically to the other end of the continuum?

23 09 2010
Dennis Newson

I always thought it was Mussolini with his: “When I hear the word culture I reach for my gun” that provided me with a template for my motto: “When I hear the word grammar I reach for my gun”, but Google and Wikipaedia have corrected me and it seems the original quotation is :“”Wenn ich Kultur höre … entsichere ich meinen Browning!” “Whenever I hear of culture… I release the safety catch of my Browning! and is uttered by a character in the play Schlageter by the German playwright and Nazi Poet Laureate, Hans Johst.

I also frequently ask myself why it is that teachers and trainers of EFL/ESOL get so very worked up about grammar. As any electronic list moderator will tell you, if there is a lull on a particular Yahoo group , you can always wake members up and generate lively discussions by simply posting a provocative statement about grammar. The messages pour in. Is it because teachers of English see that their views on the nature of grammar imply their convictions on the best way to learn and teach a language and their success in that enterprise, in the final analysis, also defines them and their competence professionally and personally? That’s just a homespun hypothesis, but something is needed to explain the passion that views on grammar generate.

I still struggle to articulate my own point of view, which I hold stubbornly and tenaciously in the face of all plausible counter arguments, but have never been able to articulate it cogently enough to start a world revolution. Here is today’s attempt to nail my theses to the door.

• According to Greenbaum (Good English and the Grammarian, Longman, 1988) there are about 12 different meanings of “grammar”. But I don’t want to play the definition game. I mean by ‘grammar’ the use of the articles, the use of prepositions, the use of tenses, IF clauses etc. You all know what I mean.

• “Grammar” is not language, though many people seem to imply that learning the grammar ( whatever “learning” and “grammar” mean *) is a vital and obligatory part of learning a language.

[ * For the record, I understand ability in a language to be a performance skill , not possession of a body of knowledge. I understand the statement: “They can speak English” as meaning something like: “They can mean what they want to mean in English producing the kind of effect on their listeners/ readers that they intend´”. ]

What can we say about those thousands of pages of descriptive grammars of English I have on a shelf behind me?

• A Comprehensive Grammar of the English Language,( Longman , both editions, 1972 & 1985) 1120 pps. + 1779 Culmative pps: 2,899 ; The Cambridge Grammar of the English Language Huddlestone and Pullam (Cambridge University Press, 2002 – “100 pounds sterling introductory offer” !!!) 1842 pps, Culmative pps: 4,741 ; A Course in Spoken English: Grammar ( Sinclair, Oxford University Press, 1972 1842 pps, Culmative pps: 6,583; Collins Cobuild English Grammar (Collins, 1990) 486 pp , Culmative pps: 7,069;A Communicative Grammar of English (Leech and Svartik, Longman, 1975) 324 pps , Culmative pps: 7,393; Cambridge Grammar of English (Carter & McCarthy, Cambridge University press, 2006 973pps , Culmative pps: 8,366; Practical English Usage (Swan, Oxford University Press, 1980, 1985) pps 658 , Culmative pps:9024 ;A Handbook of English Grammar (Zandvoort (Longman,1962,1975) pps 349 , Culmative pps: 9,373 ; Longman Grammar of Spoken and Written English (Biber, Johansson,Leech,Conrad, Finegan , Pearson, 1999 ), pps 1204 , Culmative pps: 10,577; A Reference Grammar for Students of English (Close, Longman, 1975) pps 342 , Culmative pps: 10,919; An A-Z of English Grammar and Usage (Leech Nelson, 1989) pps 575 , Culmative pps: 11,494; The Grammar Book An ESL/EFL Teacher’s Course Book ]( Celce-Murcia, Larsen-Freeman, Newberry House, 1983) pp 655 , Culmative pps:12,149; A Course in Spoken Englsih Grammar (Sinclair, Oxford University Press, 1972, pps266 , Culmative pps: 12415; English Grammar Word by Word (Chalker, Nelson, 1990, 448 , Culmative pps: 12,863.

[ This last book is quite openly a descriptive linguist’s book, written for teachers of EFL/ESL, and takes the view that teachers must have: “a conscious knowledge of the rules or the English language” though “We will be considering the means by which you can convey grammatical knowledge to your students other than explicitly lecturing on the rules, a practice that has proved its ineffectiveness over and over again.” [ Note: Grammatical knowledge, not ability to use the language communicatively.]

Of course they are of central importance when one turns to an examination, discussion, analysis of the English language system, and as an aid in devising consciousness raising exercises, when one turns to talking about the language, reflecting on what one has learned as opposed to being in the process of learning, but are they of direct use to the learner? Most of them – No. Are they of daily help as works of reference for the teacher? Like the parson’s egg – Some are good in parts. Are they of help as indicators of how languages are learned and acquired? None at all.

• Descriptive works like the books above are of interest in the first instance to linguists, applied linguists, lecturers, some but not all teachers. They are works of reference, academic, researched accounts of the language that learners (may) need to learn and teachers feel or aare told by text books they must teach, but they are not performance models, specifications of English language communicative learner skills to be mastered. At the end of the day, though only implicitly never up front, they seriously mislead teachers and learners as to the nature of language learning and acquisition. These learned books are descriptive , about language, and an attempt to make generalisations, so dangerously and misleadingly called “rules”, but their focus, by definition and design, is never ever intended to be about .the process of learning/acquisition.
• I am still going to go on releasing the safety catch of my Browning when I hear the word “grammar” because it primes me to expect to hear something with which I totally disagree about how to effectively learn a language. “Down with grammar!” say I. It has been responsible for wrong-headed teaching of languages around the globe for centuries and has seriously affected the perceptual health of millions of language teachers , learners, government officials and the general public throughout the ages.

“Throw him off the blog!” You cry. Well, hang on. Steven Krashen, for one, more moderately articulated, expresses similar thoughts. [ Book to quote from has gone temporarily missing. Can any reader of this posting kindly provide a relevant quote?]

Let me end with three quotations from Noam Chomsky, Rens Brod, the Dutch linguist and the creator of this blog. Taken together they seem to me to indicate that studying descriptive grammars is going to get learners nowhere fast.

In Language and Problems of Knowledge,MIT Press 1988 p55 Noam Chomsky wrote, referring to the formation of questions in Spanish:

“The correct rule is complex from a computational point of view and knowledge of it is not learned, it “is part of the child’s biological endowment, part of the structure of the language faculty (forming) part of the mental equipment with which the child faces the world of experience.”

Rens Brod, in Beyond Grammar: An Experience Based Model of Language, Centre for Study of Languages and Information – Stanford, California, 1998 writes: “In particular it means the knowledge of a speaker/hearer cannot be understood as a grammar, but as a statistical ensemble of language experiences that changes slightly every time a new utterance is processed.”

And Scott, in a message to the dogme list once wrote: ”Grammar ‘emerges’ – that is to say, the first stage of language learning is primarily lexical, and that grammar both accretes around high frequency words, and is distilled from high frequency memorised chunks.”

So much for the usefulness of studying “grammar” as an effective way of learning and acquiring a language.


23 09 2010
Scott Thornbury

Thanks, Dennis, for nailing your thesis on the door (of my blog)! It’s always good to hear from an unrepentant grammarophobe, even if I, personally, wouldn’t go so far as to rule out the value of some rule-based – and certainly form-focused – teaching, for certain items and for certain learners, with certain needs, and at certain stages.

In an attempt, as well, to answer Simon’s question (“So what is it you and everyone else, are suggesting I could do?”) – or, rather, not answer it (that would be too presumptuous), but at least address it, the following might seem to be, if not ideal, at least – given the knowledge-delivery paradigm we seem to be stuck in, and given the inherently linear and packaged nature of coursebooks – both plausible and practical:

1. aim for developing inituitions (‘a feel for what is right’) rather than declarative knowledge of rules and terms – this probably requires a LOT of exposure and use, combined with regular ‘grammaticality testing’ i.e. “sort these sentences into acceptable or not” (without the requirement to say why);

2. deal with grammmar after communicative engagement with language, not before – e.g. after text-based work – and deal with it as if it were just another feature of the text’s ‘texture’ (like lexis and phraseology)

3. encourage gramaticalization skills (what Diane Larsen-Freeman calls grammaring) through activities that involve ‘adding the grammar’ to what is essentially lexical, e.g. through task repetition, reconstruction activities (like dictogloss), writing after speaking, speaking after listening
etc (this is the thrust of my book ‘Uncovering Grammar’);

4. encourage retrospection on fluency activities (both productive and receptive), and indeed on whole lessons – what grammar came up (and what lexis)? and tick it off in much the same ways as bird-watchers keep a record of the birds they’ve seen

5. pattern sensitization: comb texts and transcripts for regularities, even if these are not specifically grammatical (it might just be the repetition of certain words or phrases).

6. adopt a semantic syllabus (thematic, functional, notional or a combination thereof) and make meaning the starting point of an instructional sequence, mapping the relevant forms on, as and when needed.

7. Set up activities that are consistent with this syllabus (or with the lesson’s/course’s goals as determined by the learners), and – at the point of need – provide the relevant language (including the grammar).

(This is a fairly improvised list, and I’m aware as I re-read it that there’s quite a lot of cross-over between items; also that it’s not necessarily compatible with a book, although if the book WAS just a book, full of texts, with a grammar reference section at the back, then…).

23 09 2010
Simon Greenall

Thanks Scott, this is extremely useful. I know Uncovering grammar very well but often assumed that it would be difficult to incorporate into a textbook. In fact, it wouldn’t, and in fact, there are ideas which reflect things I’ve done in textbooks.

I’m reluctant to ask too many further questions, not just because of taking up space, but because I know Dennis has got his revolver out. But here are some final thoughts about your comments:

You need to know what an item of grammar is in order to describe it, or at least be able to use some metalanguage. Is this where it strays into ‘bad’ grammar?

Do the criteria of selecting a text include some pre-emptive analysis of the grammar points which should be covered, or is it ad hoc? And if it’s the former, is it this ‘grammar syllabus’ which you reject?

I’m interested not just in procedure, but planning, and I would need to demonstrate that I’d ‘covered’ (bad word) certain structures.

I become confident that I could write and get published something which might meet all your misgivings about textbooks and grammar Mcnuggets. But you know, if it looks like a textbook, it probably is a textbook, and that may be where I’d never manage to convince you.

Thanks anyway. I wish I could crack this.

24 09 2010
Scott Thornbury

Thanks for the feedback, Simon. I confess that that list was written in haste and is NOT a blueprint for a ‘dogme’ coursebook.

What is interesting (and sad) is that a project I was involved in (years ago) to produce online materials for students and which was driven by a topic, text and task-based syllabus (the grammar distilled from the topic-appropriate texts, and recycled in the tasks wherein students either reconstructed the texts or personalised them) is now being published in print form, but the editors (impelled by their marketing people on the ground) have seen fit to reverse the sequence, starting with the grammar, thereby decontextualising – and McDonaldizing – it, and, in the process, muddying the logic of the original design, and effectively turning it into yet another grammar-driven coursebook. Ho-hum.

26 09 2010
Simon Greenall

Hi Scott and Jason,
Jason, I was hesitant to reply immediately because I didn’t want our textbook-focussed concerns to get in the way of our host’s Grammar Mcnuggets blog.

There have been practical reasons why the skills + grammar strands for adult/young adults have been separated in Japan especially, Korea and Taiwan: these strands have been traditionally taught in separate classes by different teachers – please correct me if this is no longer the case. This has been referred to as a so-called ‘four skills approach’ and is especially common in classes taught in American English.

In Chinese universities, there’s a marginally more ‘integrated approach’, in that there are teachers of reading and writing classes, and teachers of listening and speaking classes, using two separate books (not four or five) even though the primary, junior high and senior high are completely integrated within the same textbook/same class/same teacher.

In Europe and Latin America, it’s more common to find an ‘integrated approach’, where a multi-syllabus course design is taught by the same teachers and usually in the same textbook.

Jason, this may be why your ‘four skills’ series has been popular in East Asia to start, then in other Asian countries. I admit I’m both delighted and interested to learn that it has also been successful in the rest of the world.

In the late 80s publishers and writers were all involved in skills series, which were either bought as class sets or, judging by my statements over the last twenty years from the Authors Licensing and Copyright Service, largely photocopied. Gradually, teaching and learning were becoming dominated by the mega ‘integrated approach’ one-size-fits-all coursebooks.

Scott, every comment you make about why the marketing people didn’t want to go further with your material allows me to understand better the point you’re making, which is actually more constructive than the starting point for this discussion, in which textbooks, publishers and textbook writers were held responsible for grammar Mcnuggets .

Forgive me if I misrepresent you but I think you’re saying:
– the sequence in which you present the syllabus strands is critical
– starting with grammar decontextualizes it.
– focus on form ‘is arguably necessary for language development’
– you’re not advocating a no-grammar approach

There’s absolutely nothing I disagree with here, but perhaps more important, I find it extremely helpful that you should clarify things in this very focussed way.
My remaining interest is whether we should have a prescriptive grammar syllabus which needs to be ‘covered’ (sorry, I know that’s a loaded term). I’m overwhelmed by the use of the term for schools curriculum material Scope & Sequence (usually a ministry term, and adopted by publishers) and I still don’t know if the enlightened but apparently ad hoc approach to grammar you may be suggesting will be appropriate for learners who are starting their English in junior and senior high schools, and for their teachers, their teacher advisors, their parents and the ministry. I really wish it was, but let’s be realistic.

By the way, Jason yes we did talk about some of this during the discussion on YLTSIG which I moderated, and about which I wrote a report which you can find in the archives.


26 09 2010
Scott Thornbury

In haste, Simon (my flight is being called), yes, those four bullet-points (in your post) accurately represent my position, although the sequence is ‘critical’ only in the sense that whatever you start with (e.g. grammar) tends to dominate the actual classroom teaching process, and hence is at risk of subverting whatever good intentions the materials writer might have had – wouldn’t you agree? This is especially the case where the exams are grammar-focused and where the washback, therefore, is less washback than tsunami.

28 09 2010

If you ask me, you’re lucky that McDonald’s hasn’t sued you.

28 09 2010
Scott Thornbury

Yes, I live in terror of the day! 😉

28 09 2010
English Raven

Scott, your post here has inspired two related posts on my own blog 🙂

Grammar Grapple:

Wizard English Grids for Emergent Language Development:

To some extent those posts are at odds with each other (as I can at times be at odds with myself about grammar teaching, based on my own experience as a learner), but in any case, thank you for starting the juices, so to speak, that got me thinking about these issues in greater depth.


– Jason

14 05 2014

Great article.

I’m a huge fan of “The Lexical Approach” and believe corpus linguistics is the way to go (co-locations are key!).

The description of grammar co-existing within a system may be the best explanation of the true nature of grammar I’ve seen.

17 06 2020
Joe Bonner

While doing my Med. TESOL, I chose to revolve all my papers around grammar. I just find it way too hard to put all the ideas on teaching it into one take away Meta Grammar instruction Mcnugget. Ellis, Larsen-Freeman and Hinkel and many more all have written so much that a simple post doesn’t allow for a true recount of the best way to approach it. It all depends on your context.

There is a cline though. At one end, you can make a form, meaning and use clear, or you can make it harder for students to focus by showing how that point exists in the rest of the language around it. Time and space are finite in books and lessons! And we also must consider student expectations and not rocking their boat… We are paid to keep them coming back. Management shoots down well researched but newer methods too…

My own conclusions lead me to build my own learner corpus, with which I can see what my students are not doing collectively in the tasks they are attempting and can compare that to students on the same level with similar goals.

From there, I found working on complexity far more rewarding than on error correction.

For anyone interested in the research I am sharing, or would like to contribute to my own corpus research of the spaces between learner and NES corpora, grammar and vocabulary, drop a comment at https://englishgrammar.pro/

Love your work Scott


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: