The English-Learning and Languages Review Homepage



Should Language Teachers Correct Grammar Mistakes?

Behaviourism in the global English-teaching industry.


Amorey Gethin



In the Internet TESL Journal, 10, 1–18 (Gray 2004), Ronald Gray argues that in the teaching of foreign languages, correcting grammar mistakes in students’ writing is useless. In support of his case he cites many studies of the effect on students’ grammatical accuracy of various methods of correcting, or of no correcting at all. In an article which Gray calls seminal, John Truscott wrote “Often a student will repeat the same mistake over and over again, even after being corrected many times… the teacher should conclude that correction simply is not effective.” (Truscott 1996, p.341) This is true, as I know from several decades of experience. True, that is, and only, if the teacher does nothing about it.

Gray and the authors he quotes discuss at some length what teachers do, or should or should not do, but not what students should do. This is a sort of behaviourism: provide a stimulus and see what response you get. It is the fault at the heart of the whole world-wide teaching of English and other languages. The emphasis is all on teaching, and the academic research is into the effects of various pedagogies, when we should be concerned instead with learning.

Whenever I had a new class, the first thing I told them was that they had to learn how to learn for themselves. I spoke about the basic principles of how languages work. I warned of the dangers of the translation mentality. I offered them the maxims: Remembering the answer is not the problem; the problem is remembering the problem and Never assume the language works in the same way as your own; assume it is different until you discover otherwise. I corrected every mistake in their written work, with full explanations. I told them only to write what they were certain was correct, or, if they were not certain, to insert a mark to show this. I told them to make a loose-leaf list of their mistakes, note against each mistake how many times they made it, and consult the list when they wrote. And I insisted that if they were not constantly asking me questions, they were not learning properly. By getting students to work like this I had exceptional success in helping them to pass their EFL exams. I have written at some length on all these matters in Gethin & Gunnemark 1996 (pp.10-11, 17-24, 149-56, 198-99, 201).

Truscott does indeed go on to discuss the behaviour of students: “…much of L2 grammatical learning follows natural orders…problems can arise when instructional sequences are inconsistent with those orders” (p.344) and “syntactic, morphological and lexical knowledge are acquired in different manners” (p.343). Thus, although we are not discussing children learning a foreign language ‘naturally’, Truscott persists in treating students like Pavlovian dogs, objects to be manipulated according to very uncertain psychological findings, rather than free agents who can deliberately control their learning themselves.

Truscott also complains that in any case we don’t know what English grammar really looks like. “The best understanding of grammar now available is provided by current linguistic theories. But even the best theories are …incomplete …changing …inconsistent with one another.” (p.350) How right he is on that last point. And the academic accounts of the grammar of English are full of bizarreries. To mention just a few: that of Pullum, who denies that the suffix –ing has any meaning, which would signify that “He is asking questions” could mean the same as “He is asked questions” (Geoffrey Pullum in a letter to me of April 9, 1992: “…I claim that –ing has no meaning at all, it is just a meaningless ending used for a number of different grammatical purposes in English inflection and derivation…”; or Smith and Wilson, whose faith in the Chomskyan linguistics fairy-tale of wh- movement led them to concoct a fanciful fiction as a rule for the contraction of is (Gethin 1990, pp.44-48, Gethin 1999, pp.12-14, and ゲッシン2004, pp.38-42); or Quirk et al, whose accounts of the definite article and of some and any are a mass of convoluted error (Gethin 1990, pp.78-89 and at It is unlikely that many of the academic grammars of English will be of any practical use to anyone. Meanwhile, though, there are plenty of less pretentious descriptions of English grammar that are adequate for practical purposes. Mistakes still tend to be repeated in these grammars, it is true, and I have tried to correct a few of these.

Truscott and Gray emphasize that students want and expect to be corrected, yet maintain that correction can actually be harmful. People do not like being told repeatedly that they are making mistakes; their motivation is weakened. The proper response to this self-contradiction would be to gently confront students with their irrationality and explain that they must take responsibility for their own improvement. But Truscott and Gray say that time spent on grammar corrections is time not spent on “more important matters” such as the content, organization and logical development of arguments, of compositions.

I believe following such advice would be doing a great disservice to students. No training in essay construction is going to stop students writing “I am here since ten days” or “I look forward to see you”. To see how absurd the anti-correction argument is, simply consider that no polyglot would ever ask for their mistakes not to be corrected. Moreover, I have always deplored the inclusion of essay construction as one of the criteria for judging examinees’ compositions. It is very unfair. Essay writing and mastering a foreign language are two totally different skills.

If teachers really want to help their students, they will say to them: only come to one lesson a week instead of fifteen or twenty, and come one by one, and spend most of your time doing real learning on your own. But this is in practice virtually impossible for teachers to say. The global English-teaching industry is an unspoken collusion between teaching institutions, university language departments and course book publishers to avoid any honest and open debate on the subject. That would be a serious threat to the interests of all three. (Gethin 1997, and at )




Gethin, A. (1990). Antilinguistics: a critical assessment of modern
linguistic theory and practice
. Oxford: Intellect

Gethin, A. & Gunnemark, E.V. (1996). The art and science of learning
. Oxford: Intellect

Gethin, A. (1997). Learning the world language: today and tomorrow.
English Today, 49, 42-46.

Gethin, A. (1999). Language and thought: a rational enquiry into their
nature and relationship.
Exeter: Intellect, and a Japanese edition:
ゲッシン, エイモリー (2004). 言語 思考 権威. 未知谷

Gray, R. (2004). Grammar correction in ESL/EFL writing classes may
not be effective.
Internet TESL Journal, 10, 1–18.

Quirk, R., Greenbaum, S., Leech, G. & Svartvik, J. (1985). A
grammar of the English language. Longman

Smith, N. & Wilson, D. (1979). Modern linguistics: the results of
Chomsky’s revolution
. Penguin Books

Truscott, John. (1996). The case against grammar correction in L2
writing classes. Language Learning, 46:2, 327-369.


New layout 20 April 2010


The English-Learning and Languages Review and its individual contributors assert their Copyright on all the material published in it. Nevertheless, the Review gives permission for unlimited reproduction of the piece below, Should Language Teachers Correct Grammar Mistakes?, or parts of it, on condition that: (1) acknowledgement is made of the source (2) no changes are made except with the Review’s permission (3) no restraint is imposed on further reproduction of the material (4) copies are not sold



The English-Learning and Languages Review Homepage