The Oxford TEFL Blog

What are Grammar McNuggets? Scott Thornbury explains

Scott Thornbury grammar tense

Scott Thornbury is a guest tutor on our Trinity DipTESOL course and he has also taught an online MA TESOL program for The New School in New York. His latest book is Scott Thornbury’s 101 Grammar Questions (Cambridge University Press). In this blog post he explains more about the term ‘Grammar McNuggets’ which he coined over 20 years ago.

Twenty years ago, at a conference in Dublin, I coined the term ‘grammar McNuggets.’ In doing so, I wanted to capture the way that coursebooks deliver grammar in tasty, bite-sized morsels. To my way of thinking, this compartmentalization of language (an essentially shapeless and fluid phenomenon) reflected the way fast-food chains package, market and deliver their products – not so much as real food but as a simulation of real food. Just as there is no part of a real chicken that is a ‘nugget’, nuggets of grammar are equally artificial. Terms like ‘present perfect continuous’ or the ‘third conditional’ are the labels that linguists attach, after the event, to the products of language in use, but they don’t tell us much about how such language is generated and processed. Language, I argued, happens less as particles than as waves.

What’s more, the commodification of grammar, in the form of grammar McNuggets, seemed to me to perpetuate the prevailing ‘transmissive’ model of education (not just in language teaching but across the curriculum) that in turn served the interests of publishers and examination bodies. As Rutherford (1987, p. 157) had earlier argued, such entities, being ‘manipulable’,

can easily be ordered, grouped, combined, tabulated, indexed, etc. , for putting a grammatical stamp on a set of learning materials. Grammatical content in this sense is thus the point of departure for syllabus compilation, as well as its dénouement.

In this fashion, teachers are able to claim that ‘we did the present perfect’ in much the same way as tourists claim to have ‘done Istanbul’.

As ‘discrete items’ of language, grammar McNuggets are typically delivered by means of the PPP instructional model , where presentation shares semantic space with ‘a present’ (as in ‘I’m going to give you a present’), while production conjures associations with production lines and conveyor belts. This is what the Brazilian educator, Paulo Freire, called a ‘banking model’ of education, where education is construed as an act of depositing

in which the students are the depositories and the teacher is the depositor. Instead of communicating, the teacher issues communiqués and makes deposits which the students patiently receive, memorize, and repeat’ (1970, p. 53).

What is deposited are, of course, grammar McNuggets.

Twenty years on, the atomization, commodification and consumption of grammar grows apace –in parallel with the way that other commercial products have increasingly been granularized – think Nespresso capsules or laundry detergent pods.  As a case in point, Pearson is now marketing a ‘modular grammar course powered by Nearpod’, whereby, in the words of a promotional video, ‘instruction is delivered in a micro-learning format, allowing students to learn in small, digestible pieces.[1]

But do learners really learn ‘in small digestible pieces’? Or, more to the point, do these ‘small, digestible pieces’ ever cohere into communicative fluency? Is proficiency in a language the sum total of a steady and incremental accumulation of ‘small digestible pieces’?

If it were so, then it should be relatively easy to program machines to teach languages. Yet, despite exponential advances in information technologies, the promise of ‘programmed instruction’, first mooted in the middle of the last century, has yet to be realized. For all its phenomenal success, Duolingo (‘small digestible pieces’ to the power of several thousand) does not reach the parts that real language use does. As Chris Brumfit (2001 p. 12) put it:

We may learn the tokens of language formally, but we learn the system by using it through reading or writing, or conversing.

Sceptics might respond that this is exactly what the PPP model aims to achieve: teaching the tokens (words, structures), then putting them to use.  I would argue that, for a start, the range and utility of the tokens that are traditionally taught (the ‘canonical’ grammar syllabus) are extremely limited. Corpus data suggests that ‘some relatively common linguistic constructions are overlooked, while some relatively rare constructions receive considerable attention’ (Biber, et al. 1994, p. 171).

But, more importantly, pre-selecting the tokens (aka grammar McNuggets) means that the practice and production stages are tightly constrained, since the aim of any such practice is the faithful reproduction of said tokens, irrespective of the learner’s current developmental level or communicative needs. And to insist that they use the target form in the production stage runs counter to the aim of that stage, which, as originally defined by Donn Byrne, the architect of PPP, is ‘free expression’ (1976, p. 2). It’s tantamount to telling the learner: ‘You can say anything you like but you have to use the third conditional.’ As Dave Willis (1996, p. 47) tirelessly argued,

There is, it seems, little evidence to suggest that the PPP paradigm works the way it is supposed to work. It does not, as is claimed, lead in easy stages from presentation to mastery. At the same time, as a paradigm which focuses throughout on conformity it leaves little room for communicative language use.

In other words, language learning is not the ingestion of tasty nuggets of grammar deposited by the teacher and delivered, by means of a conveyor belt,  to the learner.

So, is there an alternative to a way of teaching that isn’t premised on this mechanical, conveyor belt model? Here are ten possibilities (which would probably work best in combination rather than individually):

  1. Encourage massive exposure to real language in use, through extensive reading and listening (easier nowadays with the internet), with the aim of developing a ‘feel’ for  what is accurate and idiomatic, and bypassing the need for formal rules;
  2. Deal with grammar after communicative engagement with language, 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) through identification and classification tasks, e.g. ‘Find all the examples of words ending in –ing, and sort them into groups’;
  3. Related to the above, sensitize learners to the way language is patterned, e.g. ask them to comb through written texts and transcripts of spoken language, looking for regularities, even if these are not specifically grammatical (it might just be the repetition of certain words or phrases, or of certain constructions, such as stacked nouns: speech production disorder; open-class content words; or intransitive verbs: developed, emerged, dawned, happened…;
  4. In the same spirit, use audio recordings of learners doing speaking tasks, or even recordings of whole lessons, to encourage retrospection –  ‘What grammar came up?’ – and have learners tick items off from a checklist, in much the same way that bird-watchers keep a record of the birds they’ve seen. (In online learning the recording of whole lessons is dead easy);
  5. Collect errors in learner output (spoken and written), anonymize them, and redistribute them for peer correction and analysis;
  6. Encourage ‘grammaticalization’ skills  (Batstone 1994) through activities that involve ‘adding the grammar’ to what is initially processed as essentially lexical, e.g. through task repetition, reconstruction activities (like dictogloss), writing after speaking, speaking after listening;
  7. Adopt a semantic syllabus (thematic, functional, notional or a combination thereof) rather than a structural one, and make meaning the starting point of an instructional sequence, e.g. through some kind of goal-oriented task (‘Find five concerns about the future that you share with the others in your group’) , supplying the relevant forms,  ‘at the point of need’ and ‘just in time’, and drawing attention to them through boardwork, repetition, review.
  8. At the very least, extend the range and variety of ‘bite-sized chunks’ that you teach to include high-frequency phrases, discourse markers, and the formulaic features of spoken grammar (of the kind I focus on in my session on the Oxford House Dip TESOL course): you know what I mean, that kind of thing, so anyway, …or something.
  9. Present traditional grammar not as rules, but as phrasal ‘exemplars’ (Where’ve you been? It’s going to rain, etc),  or sentence-frames (Have you ever…? There used to be…; etc) embedded into dialogues, which are practised and memorized, and from which, over time, learners will ‘extract’ the grammar (as seems to happen in first language acquisition);
  10. ‘Flip’ the classroom, assigning the explicit study and practice of grammar McNuggets as a homework task, reserving face-to-face classroom time for more interactive and communicative activities.

In the meantime, the durability of the grammar syllabus, and the McNuggets of which it is composed, suggests that, in most contexts, the above approaches might have to work alongside, or be folded into, a traditional syllabus. That might not be such a bad thing, since a two-pronged approach to language learning (particles AND waves) might be the best way of addressing its inherent complexity and variability.

If you are thinking about advancing your career in ELT, and you would like to receive training from Scott (one of our guest tutors), you could consider our internationally-recognised Trinity DipTESOL course.

Next course starts: October 12th 2020.


Batstone, R. (1994). ‘Product & process: grammar in the second language classroom’. In Bygate, et al. (eds) Grammar and the language teacher. London: Prentice Hall.

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

Brumfit, C. (2001). Individual freedom in language teaching. Oxford: Oxford University Press.

Byrne, D. (1976). Teaching Oral English. Harlow: Longman.

Freire, P. (1970). Pedagogy of the Oppressed. Harmondsworth: Penguin

Rutherford, W. (1987). Second language Grammar: learning and teaching. Harlow: Longman,

Willis, D. (1996). ‘Accuracy, fluency and conformity.’ In Willis, J. & Willis, D. (eds) Challenge and Change in Language Teaching. Oxford: Heinemann.


Comments are closed.