To grammar, or not to grammar? by Kendall Peet

There is currently a debate raging concerning the place of grammar within the EFL/ESL curriculum. On one side of the fence, arguing that grammar is a fundamental component- perhaps the fundamental component- of any syllabus, stand the ESL publishers, the authors of the published student texts, the stolid linguists, and a handful of prominent published individuals, such as Harmer, Sinclair, and Tonkyn. On the other side of the fence, arguing against the use of "packaged language" textbooks, with a grammar based linear syllabus, stand a growing body of weighty figures, such as Allwright, Lewis, Underhill, and Thornbury, who are joined by a not insignificant number of teachers in the field experimenting with different teaching methods. It is not within the scope of this article to present a comprehensive history or indeed a current account of the arguments for or against a grammar-based syllabus, but rather the purpose of this article is first to examine the argument against grammar-centred teaching, and then to look at practical alternatives, suggested by Scott Thornbury in Uncovering Grammar, that can be tested by teachers in the best interests of teacher development.

The Argument against grammar-centred teaching

In the text Uncovering Grammar, and in a series of articles, Scott Thornbury puts forward a convincing argument against the use of pre-packaged, grammar-based textbooks as the central means to teach English as a foreign or second language to students(1). He bases his argument partly on research into first language acquisition, stating that, in line with Lewis, language is first learnt in "prefabricated chunks," and that there is a natural progression from lexis to grammar, but that grammatical knowledge cannot be applied until the learner has, as Lewis writes, "a sufficiently large mental lexicon"(2). He also argues that grammar-based lessons do not lead to oral fluency, and it is oral fluency that the majority of students want most. In effect, what Thornbury is saying is that language is acquired, rather than learnt, and in doing so is reviving, in part, ideas raised by Krashen, Allwrigtht, and Prabhu, taking a somewhat Humanistic Approach, whilst at the same time supporting the limited use of relatively new theories such as TBL (task-based learning), LBT (learner-based teaching, developed by Campbell and Kryszewska), and The Lexical Approach (developed by Michael Lewis)(3). In arguing that language is acquired, rather than learnt, Thornbury is arguing for a teaching model based on a process, being the process of "emerging grammar", and not on the traditional hierarchical model of transmission.

Traditional linguists, typically university professors, and authors such as Sinclair, claim that "those who teach languages depend on those who describe them," and in so doing assert the validity of the hierarchical top-down, transmission model of education(4). Thornbury, however, rejects this claim and other such claims, which effectively proclaim "the return of grammar to the centre of language teaching and learning"(5) , fundamentally on the basis that they originate from interested parties, which is to say, they come from parties which have a financial interest in seeing the return of grammar to centre stage. He accuses the publishing houses of perverting the process of education and highlights the role that authors such as John Soars play in trivialising learning to a marketable product(6). He therefore, rejects Jeremy Harmer's view that course texts offer a "coherent syllabus… [which are] the result of many years of experience and …much research and discussion"(7), concurring with Allwright(8) , who also challenged the hegemony of gobalised coursebooks, in favouring a less ambitious text, possibly something locally produced, or, taking it a step further, moves toward LBT(9) and Underhill(10) (advocates of the material-free teaching) in advocating a materials-light approach. He argues that grammar-centred lessons divert classroom language from discourse to metadiscourse, and that coursebooks simulate rather than stimulate. In sum, Thornbury is saying that teachers need to step outside the confines of the coursebook in order to provide real communicative opportunities for learners.


Is there an alternative to a coursebook-focused, grammar-centred lesson?

In short, the answer is yes. As already mentioned, there is growing support for the idea that teaching is most effective, that learners learn best, when the communicative aspect is put before grammar. As a result, there are now many schools of thought and methodologies offering teachers an alternative to coursebook-based teaching. To highlight just a few, there is Berlitz, which continues to use the Direct Method of teaching. There is the Natural Approach developed by Krashen, based on Chomsky's ideas regarding L1 acquisition(11). There is TBL, that has evolved out the work done by Allwright and Prabhu.There are the various Humanistic Approaches, again based on Chomsky's ideas, including Suggestopaedia (developed by Lozanov) and TPR (Total Physical Response, developed by Asher)(12). There is the Lexical Approach (developed by Michael Lewis), which has gained a lot of support within the industry recently. There is LBT (Learner-based Teaching), there is the material-free approach advocated by Adrian Underhill, and finally there is the Process Teaching approach presented by Thornbury in Uncovering Grammar. Each one of the approaches mentioned, and there are others, is worthy of extended elaboration, however, in this article we shall focus on PT (Process Teaching)(13).

What is PT?

First and foremost, PT is a communicative method. PT places primary importance on the conversation and direct communication that evolves from foregrounding the inner life of both the learner and the teacher(14). The acquisition of language is, therefore, viewed more as a by-product of real communication, which is to say, PT "assumes that grammar is a kind of organic process that, in the right conditions, grows of its own accord and in its own mysterious way"(15). It is the role of the teacher to provide these conditions. In particular, Thornbury lists 10 fundamental rules that are to be used not so much as rules, but rather as guidelines for Process Teaching(16):

1. Teaching should be done using only the resources that teachers and students bring to the classroom - i.e. themselves - and whatever happens to be in the classroom.

2. No recorded listening material should be introduced into the classroom: the source of all "listening" activities should be the students and teacher themselves. The only recorded material that is used should be that made in the classroom itself, e.g. recording students in pair or group work for later re-play and analysis.

3. The teacher must sit down at all times that the students are seated, except when monitoring group or pair work (and even then it may be best to pull up a chair). In small classes, teaching should take place around a single table.

4. All the teacher's questions must be "real" questions (such as "Do you like oysters?" Or "What did you do on Saturday?"), not "display" questions (such as "What's the past of the verb to go?" or "Is there a clock on the wall?")

5. Slavish adherence to a method (such as audiolingualism, Silent Way, TPR, task-based learning, suggestopaedia) is unacceptable.

6. A pre-planned syllabus of pre-selected and graded grammar items is forbidden. Any grammar that is the focus of instruction should emerge from the lesson content, not dictate it.

7. Topics that are generated by the students themselves must be given priority over any other input.

8. Grading of students into different levels is disallowed: students should be free to join the class that they feel most comfortable in, whether for social reasons, or for reasons of mutual intelligibility, or both. As in other forms of human social interaction, diversity should be accommodated, even welcomed, but not proscribed.

9. The criteria and administration of any testing procedures must be negotiated with the learners.

10. Teachers themselves will be evaluated according to only one criterion: that they are not boring. Thus, in contrast to the Direct Method, he does not exclude grammar, but rather he argues against the Direct Method and the Natural Approach in writing that "A diet of nothing but unrehearsed fluency activities…may make learners over-reliant on lexical processes at the expense of developing their grammatical competence"(17). PT is therefore presented as a balanced approach, which recognises the importance of both fluency and accuracy activities, with grammar emerging organically from the language. Or to put it another way, Process Teaching is effectively Jeremy Harmer's Communicative Approach, but with emerging grammar, rather than grammar as an end product, and less emphasis placed on course texts.

What then is the role of the teacher in PT?

PT states that the role of the teacher is to:

  • Provide comprehensible input
  • Facilitate interaction
  • Facilitate item learning
  • Facilitate pattern detection
  • Provide output opportunities
  • Provide feedback
  • Motivate learners

Looking at this list, the role of a PT teacher does not appear radically different to that of say a PPP teacher. However, as Rule 2 suggests, there is greater emphasis placed upon TTT. Furthermore, as Rule 4 suggests, the nature of communication within the classroom is also quite different. During conversation, the teachers seeks to provide "scaffolding," by means of correcting, prompting, recasting, by supplying needed language, and by making comments, to exploit any learner small talk. In this respect, the teacher is more a facilitator than "A teacher!" However, as mentioned, at some stage the teacher will need to highlight certain language points that emerge from the conversation. The teacher will also need to help the learner to develop effective learning strategies with regard to recording and recycling the language that emerges, so that the items "start to cohere and fuse into generative patterns"(18). One particular learning strategy, which the teacher models, is the practice of "noticing" patterns present in language. In time, the learners will also begin to "notice" patterns and in so doing will take an important step toward becoming autonomous learners. Another important aspect of Process Teaching is feedback, both positive and negative. Thornbury argues that teachers need to highlight the gap that exists between the learner's language and the target language, the gap between effective and less than effective communication, in order to motivate learners to internally restructure their language.

To highlight more clearly how the role of a traditional teacher differs to that of PT facilitator, it is perhaps most useful to look at the changing assumptions(19).

I am:    
A Teacher   A Helper/Facilitator
We focus on:    
What   How
Product   Process
The topic   The way we work on the topic
I know it/they don't   We are on the same side of the learning fence
We do it in:    
A Classroom   A Workshop
One implicit message is:    
Be correct   See what happens
As a result mistakes are:    
To be avoided   Welcome; part of the process
And I ask questions:    
To elicit a right answer   To learn something myself
So I tend to:    
Listen to their language   Listen to their process as well as their language
I treat learning as being:    
From outside in   From inside out
Starting from the book   Starting from them
I view the syllabus as:    
An Outer Syllabus   An Inner Syllabus
Linear   Organic
Imposed   Determined by the learners
Activities are:    
Decided by me/the book   Negotiated & naturally arising
Time is:    
Mine   Ours
This means:    
I can get impatient   I respect the time it takes

As you can see, one of the most significant differences is that a PT teacher is a member of the class that shares in the learning experience. The teacher does not hold a divine position in the order of knowledge, preaching to his subjects from the towering pulpit. No, instead, learning is regarded as an interactive, living, breathing process, that seeks to broaden and strengthen the highway of communication through the process of communication. The teacher is there to facilitate the emergence of language, rather than to instruct or teach.

Problems inherent in the process of implementing the PT approach

It should be noted that any teaching model that deviates considerably from the traditional top-down transmission model of teaching is going to face problems. One major hurdle can be the learners themselves, who often come to class with preconceptions about the learning process. The idea of learning grammar and writing copious amounts match up in the minds of many learners to the model of a good lesson: I have struck this attitude myself on many occasions, particularly with the older adult students. There can also be a problem in matching this communicative style of learning to the particular learner, as not all learners are best suited to a communicative method. There can be syllabus constraints, owing to exams, and then there are also the constraints which are placed upon the teacher in having to take full responsibility for the syllabus, which is normally left to the coursebook in conventional teaching. This final point is of particular importance for less experienced teachers. I initially found coursebooks extremely useful and cannot imagine how I could have taught without them. Underhill likewise emphasises the important role that published materials, particularly coursebooks, play in the the early stages of teacher development. As with most things learnt, for example the piano or driving a car, it is only after one has received proper training that one has the luxury of making ones own way unassisted. Therefore, in view of these arguments, like Campbell and Kryszewska, Thornbury conceded that a compromise must be struck in regard to the use of a course textbook. The compromise, in keeping with the emergent view of grammar, therefore stipulates that teachers should:

  • Work from texts, tasks, and topics rather than from a structural syllabus.
  • Generate language and then look for items and patterns.
  • Talk to learners and scaffold their language.

PT Activities

So what type of activities fall within the confines of PT? The answer quite simply is any activity that is about something, that is topical, that draws on real language, real communication in a way that activates interest in the learner and as a result produces a lot of rough language, from which the grammar can emerge through the process of noticing: in brief, activities such as free discussion, questionnaires and surveys, teacher anecdotes, student stories, CLL (Community Language Learning) activities, and paper conversations (for a more comprehensive list refer to the Appendix 3). When selecting an activity, teachers might do well to bear in mind the belief that "the best classroom activities are those that incorporate elements of real life language"(20) such as making plans, arguing a point, negotiating, reaching a consensus, or finalising an arrangement, etc. The question to ask is: How will this activity and the inherent language benefit the learner?


The world is undergoing a period of rapid change. In the 1915, the modernist poet, Ezra Pound, declared the need to "Make it new"(21). Today, in this postmodernist age, where pastiche reigns supreme, the call going out is to Make it real. Is process teaching the way ahead? I leave you to answer that question. What I will say is that as teachers we continually need to redefine our role within the classroom, which can only be achieved by experimenting with different methods of teaching. I am not sure that there is, or ever will be, a single best teaching method, for the simple reason that there are many types of learners with differing learning preferences. What I do believe, however, is that every teaching model or approach has something to offer and that what is most important is that teachers seek to rewrite themselves again and again, for it is in rewriting yourself that you find you have something new to say, something that you can say with enthusiasm and conviction; and if the teacher is enthusiastic and motivated, it is likely that the students will be too.

1 The articles referred listed in the Bibliography.
2 Thornbury, S. 2001. Uncovering Grammar. pp. 16, 17
Lewis, M. 1997. The Lexical Approach. p.15.
3 If you are unfamiliar with the theories presented by the authors mentioned, please refer to the bibliography.
4 Sinclair, J. 1997. "Corpus evidence in language description". pp. 27-39
5 Tonkyn, A. I994. "Introduction: grammar and the language teacher". pp 1-14
6 Thornbury, S. 1998. "Grammar, Power and Bottled Water". IATEFL Newsletter, 140. pp 19-20.
7 Thornbury, S. & Meddings, L. July, 2001. "Coursebooks: The Roaring in the Chimney". MET, Vol. 10, No. 3.
8 Allwright, 1981, 1990. "What do we want teaching materials for?" pp. 131-147.
9 Campbell, C. & Kryszewska, H. 1992. Learner-based teaching. p. 11
10 Underhill, A. "Teaching without a coursebook".
11 Lowe, M. July, 2002. "Stepping out of Chomsky's shadow". Lowe explains that Stephen Krashen founded the Natural Approach
based on the idea that adults acquire L2 in essentially the same way as children acquire L1; a view that is now strongly questioned
by many linguists, including Lowe. Lowe also writes that Chomsky "helped to create the climate in which MET's defender of
coursebook use (Jeremy Harmer) has a defensive air, while the attacker (Scott Thornbury) is felt to be on the side of angels. p. 49
12 Lowe, M. July, 2002. ibid. Lowe argues that if Chomsky's ideas are found to be flawed, then the Humanistic Approaches will lose
much of their theoretical justification. p. 49
13 Thornbury has not yet named his approach, therefore, for convenience purposes alone, I shall henceforth refer to it as PT (process
teaching), which I have derived from the title in chapter five, "Process teaching", from the fact that he returns to the idea of process again and again throughout Uncovering Grammar, and from the following quote, "I will argue that grammar is better considered as a dynamic process", found on p. vi..
14 Thornbury, S. Feb, 2001. "Teaching Unplugged". p. 3
15 Thornbury, S. 2001. Uncovering Grammar. p. vi
16 Thornbury, S. Feb, 2001. "Teaching Unplugged". p. 4
17 Thornbury, S. 2001. Uncovering Grammar. p. 20
18 Thornbury,S. Ibid, p64

Thornbury, S. Ibid. p. 64
20 Underhill, A. "Teaching without a course book". p. 3
21 Thornbury, S & Meddings, L. "Dogme and the coursebook". MET.


Kendall Peet has taught in Thailand, South Korea, and Turkey, and is currently teaching at FIBSB in Bucharest. He has completed the RSA DELTA and is presently completing his MA in Applied Linguistics. His key interests include teaching academic writing and developing a needs-based (learner-led) approach that encourages greater learner autonomy. Kendall can be contacted at:

To the original article

Back to the articles index

Tips & Newsletter Sign up —  Current Tip —  Past Tips 
Train with us — Online Development Courses — Lesson Plan Index 
Phonology —  English-To-Go Lesson  Articles Books
 Links —  Contact — Advertising — Web Hosting — Front page

Copyright 2000-2014© Developing