Scott Thornbury

Error correction – Scott Thornbury

Scott Thornbury

What are errors & how should we  deal with them in our classes?

– Scott Thornbury

When I first started teaching the answer to these two questions was clear and unproblematic. What are errors? They are any departure from standard English. How should we deal with them? We should correct them lest they become ‘bad habits’.

Subsequently, these two questions have become the most difficult, problematic and mysterious of all questions related to language teaching.

What are errors? We simply don’t know any more. Why? Because there is no agreed upon standard by which to measure learners’ output. For a start, there are so many varieties of native speaker English (both spoken and written) that it’s impossible to decide if a sentence like ‘I can’t get no satisfaction’ is ‘wrong’ or not. On top of that, many learners are not interested in speaking ‘native speaker’ English anyway.

What should we do about errors? Research suggests that correcting errors has only an accidental effect on accuracy, and that many so-called errors (like failure to add –s to present simple third person singular verbs, as in she work) are an inevitable stage of language learning, and are extremely resistant to correction. On the other hand, if we don’t correct errors we may send out a message that accuracy doesn’t matter, which may threaten the long-term language development of our learners. Also, we need to be aware that excessive correction can be very de-motivating for many learners, while not to correct errors will make us look incompetent in the eyes of other learners.

In short, errors, and the way we handle them, are an enormous puzzle, and I would be fascinated to know how you deal with this puzzle yourself.


Connect with  our iTDi Associates, Mentors, and Faculty by joining  the iTDi Community. Sign Up For A Free iTDi Account to create your profile and get immediate access to our social forums and trial lessons from our English For Teachers and Teacher Development courses.

Like what we do? Become an iTDi Patron.
Your support makes a difference.

Published by

Scott Thornbury

Scott is a teacher and teacher educator, with over 30 years' experience in English language teaching. He is currently Associate Professor of English Language Studies at the New School in New York, teaching on an on-line MATESOL program. His previous experience includes teaching and teacher training in Egypt, UK, Spain, and his native New Zealand. Scott’s writing credits include several award-winning books for teachers on language and methodology including The A-Z of ELT, How to Teach Grammar and Teaching Unplugged. He is series editor for the Cambridge Handbooks for Teachers (CUP) and was also the co-founder of the dogme ELT group, whose archived website, called Teaching Unplugged, can be found below. Scott currently leads a fascinating community at the popular and thought-provoking blog, A-Z of ELT blog. Scott is lead author in the iTDi Teacher Development program as well as being iTDi's Academic Director.

35 thoughts on “Error correction – Scott Thornbury”

  1. Hi Scott,

    One of the courses I teach is discussion/debate and as students allegedly learn or have learnt English online and had grammar lectures, the course is supposed to be speaking-based. I tried doing my usual dogme ‘language focus’ but it didn’t go down well, I also don’t have any boards, just a projector and a wall. Thus, I started focussing on error correction and supporting language creation but with an emphasis on helping students put across their ideas, not on saying something is wrong. Now, there’s a lot of peer support and if I correct something once and the student reoffends the others pick up on it.

    As I said, it is a discussion and it has evolved into me acting like an orchestra conductor in a way. Sometimes pair work, group work or even a debate seems appropriate, other times just a class discussion seems to work. Whatever the format, because I am interested in creating real conversation with ideas, opinions and views, however controversial, error correction is welcomed and often asked for. I’m also now the ‘last resort’ in a way after self-correction and peer.

    1. Thanks for the comment Phil. It sounds like you have created the optimal ‘learning environment’, at least insofar as errors go. When the classroom becomes a kind of ‘self-healing’ organism, with the teacher as last resort, you have a really potent content for ‘scaffolded’ risk-taking.

  2. I’ve recently came across an approach that deals with errors similar to Kolb’s experential learning cycle. It takes time, but it allows the learner to reflect on errors and avoid making them again.

        1. It’s years later, but I’m wishing for a miracle…and a link! What it makes me think of is this, anyway: which is the framework into which I tend to place the role of corrective feedback in learning.

  3. Not only is this area a huge puzzle, (spoken) errors (and their correction) are also things that a teacher has to be able to make almost instantaneous decisions about in the moment of teaching….do I correct it or not? why is it wrong, what is correct in this context/instance?, what level is the learner and how much do they know / can I ‘push’ them here? how do i correct it? now or later? how is the learner going to feel? – all these thoughts and decisions have to be answered in the moment, and that’s pretty overwhelming.
    As much as I’d like to say I have a principled approach that is based on my awareness and consideration of these issues, it often isn’t 🙁
    I might deal with errors just as I might deal with communication breakdown with other (native) speakers e.g Sorry, did you mean…? or What do you mean? at times, and at other times I make sure there is a stage of the task/lesson that deals with inaccuracies that were heard.
    I’d like to believe that this is a very eclectic approach that attempts to focus on learner needs, and the dynamics of the class 🙂

    1. Thanks for the comment Kristin. I guess we all like to think our approach to error correction is ‘principled’, but as you say, in the heat of the moment, it’s very difficult to juggle all the factors – cognitive, affective, cultural etc – that impact on how we deal with errors that arise in spontaneous talk. Perhaps that’s where teaching is more an art than a science.

      1. For me, a combination approach has always helped. Understanding learning situations and the learner mix is extremely important to the teacher. I’m sure most teachers agree that it s much easier to provide explicit feedback to younger learners. An indirect approach is better whilst handling a class of upper intermediates or adults.
        Errors that come in way of communicative competence cannot be compromised or ignored. Errors in accuracy can probably be handled in indirect manner involving the rest of the class in diagnosing and correcting the errors

  4. A colleague of mine talks a bout three types of error – accuracy, intelligibility and irritation.

    Accuracy means correcting just because it’s not correct. He feels these should be left alone, particularly if the learner’s meaning is clear.

    He mainly feels you should only correct when the error makes it difficult to understand the learner.

    Most interesting is his ‘irritation’ class. If he feels the error irritates native or advanced speakers, he corrects it.

    I like this idea but am also wary about who decides what irritates

    1. Has there been any sudies that shows when you leave an accuracy error alone (for example 3rd person, present +s) the learners actually self corrects over time through just exposure (reading & listening) to correct language from the teacher of other materials / sources?

      1. Hmmm I haven’t heard of any such studies but it’s something I’d be curious to see. Mind you, it would be very difficult to operationalize such a longitudinal study with two groups of learners.

    2. Thanks Ed – I like your (or your colleague’s) classification, particularly the ‘irritation’ category. I think we do owe it to our students to indicate the errors that are likely to alienate their interlocutors. I distinctly remember a Spanish teacher I had who told me, almost from day one, that she was driven crazy by learners saying ‘una problema’ instead of ‘un problema’. It made such an impression that I very rarely make that mistake!

  5. I agree with you Scott in saying that error correction is one of the most problematic areas of language teaching.Hundreds of articles and book chapters are written about this area of ELT.

    I am always puzzled as to wether I should correct a student’s error or not as I have many cases where this correction certainly had side-effects or counter-productive results.
    now,I always take some time in the beginning of the school year so that I know something about the personality of my students before I start directly or even indirectly correcting their errors.I have discovered- and most educators might have seen this as well-that some students are too fragile and they might react in a way too personal to the teacher/other student’s correction/feedback on their errors.

    This doesn’t mean that some errors shouldn’t be corrected.In cases where error correction hurts,I have resorted to “let it to another day” or delayed error correction;in other words,if I see that a students keeps making the same error ,I find a way to correct the error when he/she is not aware that I am referring to him/her.I sometimes feel that if the student is nor corrected on the spot,he/she wouldn’t be aware that he’s:she’s making that mistake/error.I have found this ,however,better than switching off a student once and for all.

    Another poin I would like to make is that how much correction really happens when we correct a student’s error.I am not aware of the existence of any research about if correcting an error is really what “corrects” the student’s error.I suppose that so many errors get corrected along the way without the teacher’s intervention;and this is where other language skills play a great role.I strongly believe that extended reading and writing have a major role to play in chaning a student’s misconceptions of a language item.I am getting pro comprehensible input -krashen- here;and I really agre with the hypothesis that sports Krashen “comprehensible input” theory.

    Talking about L1,how much error correction do our mothers/fathers do while we are learning our first language?I have a one year old baby now – and already another oen before him- and I have not noticed myself or his mother correcting our baby’s errors;and yet so many sentnces emerge at an early age.It might be a school fashion then to correct errors,which no one is aware that they are really corrected that way.I suppose that creating the L1 conditions in L2 or FL learning is what really needs to be taken into consideration.

    Much language learning happens exactly the same way as other learning”(s)” about life do.Children learn language by observing an older person around them,that observation involves listening to him/her and trying to emitate him/her.This might explain why so many “family/tribe” “errors” continue and resist change despite school’s correction.I am here referring to black English that might be corrected by so many teachers in school and yet it continues to evolve and distinguish itself as a distinct language.I have also witnessed cases where some of our students exchange the /l/ and /r/ sounds and they never orrect them despite our efforts and sometimes the sarcasm of other peer students.

    So many misteries remain untapped by research,and language errors continue to be a mystery.

    1. The question then arises is there a way to examine implicit focus on form affects linguistic accuracy of a group of EFL learners, written production of language proficiency in comparison with another group receiving delayed, explicit focus on form at the same level of language proficiency?.Are there any statistically significant differences between Explicit teaching and Implicit teaching when teaching the usage of grammar?

    2. Thanks, Brahim, for your comment. I think that what you say here gets to the heart of it:

      “I am not aware of the existence of any research about if correcting an error is really what “corrects” the student’s error.I suppose that so many errors get corrected along the way without the teacher’s intervention;and this is where other language skills play a great role”.

      I think it’s probably a case of both – that is to say, some errors are susceptible to the teacher’s intervention, while others just resolve themselves over time and through exposure and practice. The problem is, the constellation of ‘correctable’ errors and ‘non-correctable’ errors will very from learner to learner, so maybe – as a safe bet – we should correct more often than not, in the hope that at least some of our corrections will hit the target. Just a thought.

  6. One of my approaches is very similar to Phil Wade’s. I always encourage collaboration & wide open commumnication channels in my classes & part of this is peer – to peer error correction and improvement. Taken from your ‘unlpugged teaching’ book, I prefer to use the word unsuccessful language and improvement” rather than error or mistake. One of my favourite tasks is one from your book, where I make a list of utternaces (successful and not) and the learners must discuss each one and try to locate any unsuccesssful language, correct it and then also improve the utterance (if possible). I also find that a good method is getting the learners to produce and deliver mini-lesson & tasks on the area of language that is causing them problems – this works better though when it deals with something tangible like a piece / area of grammar.

    1. Thanks, Karl, for your comment – and thanks for mentioning that activity from ‘Teaching Unplugged’. I had a class who told me that one of the activities that they really liked was just going through a list of errors I had collected from a previous lesson, or from their homework – all anonymised – and working on correcting them. Something satisfying in getting things right, especially when there’s no threat to one’s public ‘face’.

  7. The main errors I find students all over making are: failure to add -s to the third person (as discussed previosly), incorrect preposition use, and failure to use the past simple (what i call the ‘yesterday i go’ syndrome).
    It seems to me that these errors are mostly the result of a) lack of self-awareness and b) – something which i feel is grossly overlooked – the fact that the correct form WAS NEVER TAUGHT IN THE FIRST PLACE. In this case, how can we expect our students to get it right?
    Let’s take the third person singular ‘-s’. When I studied French and German at school, a great deal of time was spent learning to conjugate the verbs. Now, in English, because only the third person singular is subject to an ending being added, students are never taught how to conjugate verbs in the present simple (or any tense for that matter). It seems it is our general practise to merely *tell* our students “oh, and don’t forget, when you’re talking about he or she, you have to put an ‘s’ at the end”. Of course, the student nods, more or less understands, BUT CRUCIALLY HASN’T PRACTISED AND THEREFORE HASN’T LEARNED IT.
    Clearly this is a mistake in our approach as it sends out a message to students that the ending doesn’t matter, that verbs don’t have endings, students of course subsequently don’t practise this hence the mistakes they make.
    Therefore, I make a point of focusing just on third person ‘s’ on a regular basis hopefully so these kind of mistakes don’t become embedded.
    I give a prompt such as ‘Right, SS A, He – Want – Question’ to which the reply should be ‘What does he want’ I then drill this with other verbs. Then I might change it to ‘Give me a question with ‘who’ and ‘know’ by which I’m trying to elicit “Who knows?” or “Who knows where i can find a good shoe shop in this town?”
    OK, so these drills are not exactly pretty, tend to be teacher-centred (although you can get them to do it in pairs as well) but they are crucial.

    1. Interesting comment, Steven. I think you’ll find that there are many teachers out there (I was one!) who have definitely TAUGHT the present simple -s, in the sense of having presented it, drilled it, practised it, corrected it – endlessly – and still students resist using it. This is fairly well attested in the literature on second language acquisition – certain grammatical structures simply resist teaching. Why?

      Take the third person -s. How many good reasons might there be for students NOT to learn it?

      1. It consists of one sound and one letter – difficult to perceive, compared to, say, the ‘-ing’ ending, which is typically learned much quicker.
      2. It’s redundant, since in English we always state the subject of a verb, even if it’s a pronoun, so we know that ‘he work’ refers to ‘he’. Why flag the third person twice?
      3. The -s ending indexes the plural in many languages, including Engish – so it seems odd to add it to a singular verb.
      4. In many Latin-based languages, the -s ending, if added to a verb, signals the 2nd person (tu hablas) not the 3rd person, so, for speakers of these languages, it is counterintuitive.
      5. In many regional versions of English, it is not used.

      So, for all these reasons, the -s ending in the third person singular present tense verbs is VERY difficult to acquire, and resists both teaching and correction.

      1. On the topic of the 3rd Person Singular -S, AKA the “little despot”! –> I focus on it most when/if I know my students are going to a US university, etc. and will have many NS interlocutors. I can’t get those “baby-talk”/”uneducated” quotes out of my mind..and want my students to avoid it having thought about them wrongly!

  8. Dear Scott,
    The term error correction can be quite a grey area as you mentioned. What is error? Are we talking in terms of grammatical errors, wrong usage of words, pronunciation errors or…..? The list can go on and on and be overwhelming.
    I usually set specific focus, rather than addressing the different errors together. For example, when working on a particular grammar tense (like the past tense), my focus would be on giving feedback on the appropriate usage of past tense (either by means of monitoring peer feedback or addressing the error as a class), whereas when working on a writing piece,I would then focus on meaningful connectivity aka cohesion in text.

    I find that a purposeful sense of focus during error feedback sessions really help. Learners themselves are able to work slowly, step by step, in areas that need improvement, rather than being bombarded with torrents of feedback that , in the end, leads them to feel flustered and demotivated.

    As Anne Lamott says “bird by bird, buddy, just take it bird by bird”…..:)


  9. Errors are wonderful windows. They give the teacher a way into the thought processes and speed bumps of language acquisition.
    Why do students make the same errors over and over?
    Is it a lack of knowledge, or is it previous knowledge?
    L1 influence can really frustrate the learning but it does show that critical thinking is happening.
    Here in Indonesia it’s quite common to hear the statement, “I have ever been to Bali.” It can be a frustrating error until a teacher(and the student) realize that the Indonesian sentence,”Saya sudah pernah ke Bali.” doesn’t translate directly into English as we shouldn’t use ever(pernah) in statements. This gives the teacher a way to make general corrections. L1 is only one of many reasons for errors.
    My long way to answer the question. Error correction is a dance, requiring the teacher to have an awareness of their surroundings and a certain nimbleness in their steps.

  10. As Wayne says, errors are a great window. However we frame what we see in terms of our understandings. If you don’t know what a car is, seeing it from that window will not shed much light on the subject. You have got to get out there! 🙂

    From my experience and understandings errors can be caused by one of the following factors..sometimes even a mix!

    – ignorance ( the person just does not know)..this needs to be sorted out by the teacher before one can know what to do next
    – lack of attention ( person knows and could self correct,but has not automated the usage yet, so if his/her attention is caught somewhere else, mistakes like that can easily happen when one is learning) – this can easily be sorted out by the teacher
    – has half an understanding, so the mistake is made by the learner trying to clarify (teacher can also help out here by establishing what is known)
    – the learner knows but does not really care if they get it wrong, so they pay insufficient attention ( or none) to get it right. Sometimes this can be seen as the learner being unwilling to pay the price to do what they know needs to be done.

    ( there may be more, but it’s getting late! 🙂 )

  11. I would venture saying that errors are an integral part of language learning. Students will keep making more or less the same mistakes despite our constant and regular correction .The whole operation is happening inside that is for learners to get it right ,the brain should “kicks in” as it has been said by Azar. Simply put, They should be aware and conscious of it to avoid it later on. And I personally noticed that in class. It is more or less like a battle and too much correction as you have mentioned is very de-motivating indeed. Teachers tend to correct too much because they are pedantic and some even trying to be perfect .Since we are talking about errors, there are some scholars who advocate zero grammar teaching .What’s the point of teaching it if students keep making the mistakes .”It is like saying we plant seeds, but not all seeds come up .Therefore, it is no use planting seeds”.

  12. I’m 100 per cent agreed with you; errors just exist and are a very challenging puzzle even for the greatest teachers. In my case, particularly, at times I correct mistakes just by showing the write way (formal instruction), others I decided my students to correct each other nicely of course, because some students tend to be so critical and rude. I also have an audio/video record of my students’ activities and show them after a while so they can correct themselves, as I wrote time ago, sometimes they just have a memory or language lapse, so I give them the opportunity to see their own mistakes and correct them. When it is about written assignments I use to highlight mistakes and talk about them in class, and to my surprise, most of the times students have already done a bit research about what and why I highlighted some things in their papers. One of my career teachers said that we should correct students’ mistakes by giving points and after having many bad points maybe they will get it and not continue to make them. I’m not so sure this is the best or even worthy of consideration way, but I do know that if you highlight many times the same things, students will notice and do something on that regard. Finally I can say that every time mistakes appear –and they certainly will, we, as teachers, will need to figure out how to deal with them the most suitable and less troublesome way.

  13. No to hurt my students, I never interrupt my students while they are speaking. After they finish, I first point out the good parts of their speech and then mention their major mistakes and explain the correct way to say it. Sometimes I ask the group what they liked about the speech or the presentation and whether they have noticed any mistakes.
    When my students write essays I always analyze mistakes in class without giving the name of the authors.

  14. In coming to teacher later in life, I think many instructors fear an error, but students have the same fear. “Oh, my, what if I make a mistake in front of all these people. How will it appear?” I tell the students that it makes us all appear the same — human. In teaching journalism and writing, I see so many broken rules. In many cases, the rules are made to be broken. But I feel that we need to be able to build with the basics before we can branch out and start inserting a fragment here and there. So I really work with the students on basics. In class today, I had turn in their lead homework. I reviewed every lead on the board. Had any of them read the work out loud, they would have heard the errors. Again and again, I asked the students what words I should/could change to make the narrative clear. After that, I asked them to rewrite the lead in class, within 10 minutes. We reviewed there too. No names. Just the words. But they learn. On one level we need mistakes and errors to learn. If we hide from them or cover them or run from them, we’re just screwed. If I make a mistake, I own up. So what? Who cares? I try my best to never defend when a student points out an error. I tell student I’m glad I am still making mistakes; it means I am still growing. I realize now that fear was the failing of many of my college professors. They were terrified of being seen as not knowing or being in error. When I was in college, I just thought they were being coy or brilliant or something, I don’t really know. Now, I realized many of them were terrified. That’s such a shame. I tell my students to leave their egos at the door. Anything that is born from ego is based on fear and only hampers progress. I too leave my ego at the door and always ask that the universe and the great scholars who came before guide me and help me stand at the threshold. It sounds silly, but it seems to work. Holy cow, I just realized this error blog could be about something else. I could be in error. Oh, well, never mind. Live and learn.

  15. I always start a course by inviting and encouraging lots of mistakes, as a sign that students will be trying to communicate in ways they have not yet perfected, which is crucial for progress.

    So after a discussion between students, I will introduce my notes saying “Thank you, lots of wonderful mistakes!”

    Later I introduce a distinction between old and new mistakes, encouraging the latter.

    With a student who can take it, I will interrupt them when they make of the serious fossilized errors we have previously identified, but otherwise leave it till later.

    Last comment: when bringing to class a list of previous errors, as Scott mentions, I find it can help to “camouflage” them by changing non-essential words. That way sensitive students aren’t sure if any given error was theirs, especially if they were all writing on a similar topic.

  16. Ha!ha! Just last Saturday my 8-9 year old language students found an error in their textbook! I was so pleased. I congratulated them and told them to cross out the wrong word and replace it with the correct one. The error (gasps!) “You ‘eat’ soup from this”. They immediately shouted out that you cannot eat soup, you drink it.

  17. Hello Scott!
    Loved hearing you today…(excuse the very long winded comment :))

    I love your comment here…”the classroom becomes a kind of ‘self-healing’ organism, with the teacher as last resort, you have a really potent content for ‘scaffolded’ risk-taking”.

    In every class I teach that is what I have always strive to achieve. I think it is important that in doing so to 1) assess who your students are, 2) what are their aims, 3) where are they coming from, and then 4) set up, from the start, an environment that welcomes this kind of learning. It requires a very light handed approach and a great deal of humor. When done right, the adult students are relaxed and open to learning, correcting some fossilized issues, and tend to retain better. The children are engaged and learning through play, as is natural developmentally.

    I think it is very important to distinguish the differences in the approaches with children vs adults. Having taught both, I can say there are great differences in them. Children come from a very different place, which is why I emphasize the importance of knowing where your students come from…not just in location, as in country, but in experience and previous knowledge as well.

    Also, having taught many different aged students, from many different countries around the world, very often at the same time, it is extremely important to be sensitive to the where they come from in creating the right environment. In employing instructional conversation, which I do 90% of the time, I think one has to be sensitive to when it is good and when perhaps a slightly different approach would be more appropriate. Understanding all the different methods of error intervention is important…and when to use which method, depending on the particular student. The key is to teach the students you have not the ones you think you should have.

    Oh, on an interesting note…someone in the discussion noted that NS are fun and NNS are serious teachers. I don’t know about that. I recently polled a class and evidently they all loved my class best because it was fun. Yet, they also confessed to another teacher that I was very strict. So, I guess it can be fun within very strict boundaries. I guess from my early developmental experience, I have found it is important to have strict boundaries. This provides a safe space for learning, playing and making mistakes. The students always understand there is a safety net so it is okay to make those mistakes. The interesting note was this was an adult class that thought I was very strict yet a lot of fun.

    Thanks Scott for, yet again, wonderful new nuggets of learning. No matter how old the conversation there is always something new! 🙂

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.