23 Comments
  • Cintia Zaitune
    Posted at 14:43h, 10 fevereiro Responder

    Great, Luiz Otávio! I had never thought about it this way, but totally agree with you! Thanks for this! 🙂

    • Luiz Otávio Barros
      Luiz Otávio Barros
      Posted at 17:35h, 10 fevereiro Responder

      Cintia,
      Thank you so much.
      Um beijo!

  • bjarne
    Posted at 18:32h, 10 fevereiro Responder

    Dear Otavio

    Once again an interesting article that hits an area ot is ohh so hard to absolute about.

    I agree with you about it being a good idea to do more immediate correction. Another reason is that maybe we not just over-estimate the emotional impact, but maybe also because we underestimate our students’ ability to apply there ability to keep a train of thought despite interruptions. I mean, when we are interrupted by a waiter or any other unforseen event in conversation most of us are able to go; “Where were we?” and navigate back to the conversation again . That is a useful skill to transfer to a foreign language situation I reckon.

    About your example of the a/b pairwork, isn´t it that delayed correction would not be ideal as you were doing controlled practice? And if the mistake is target language, the predictaed repetition of the target language would lead to repetition of the mistake. Hence, maybe we can say that target language mistakes (or expressions that may be repeted a lot because of the nature of the activity) in controlled practice should generally be corrected on the spot, while one off mistakes in freer practice can be left for delayed correction. That is if students remember.

    • Luiz Otávio Barros
      Luiz Otávio Barros
      Posted at 19:50h, 10 fevereiro Responder

      Hi, Bjarne.
      Thanks for your comments!
      Interesting point about students’ ability to deal with interruptions. Never thought about it that way!
      As to the there is / are activity, it wasn’t really controlled practice. It was one of those Penny Ur photocopiables, in which they had to describe the photos (without looking at their peers’) using a variety of structures (prepositions, adjectives, present continuous etc) including there is / are. More like quasi-communicative practice because of scope and processing demands.
      But, yes, if it’d been controlled practice, then on the spot would’ve been my natural choice.

  • Ricardo
    Ricardo
    Posted at 18:45h, 10 fevereiro Responder

    I’m happy you are back to blogging, Luiz! I feel the same way about delayed correction and have been trying to correct students on the spot more often (not sure to what extent I have succeeded, though).
    I don’t want to make any excuses, but do you feel on the spot correction is at all connected to how tired a teacher is? I feel that at the beginning of the term (when I’m excited and full of energy) I end up correcting people more than I do at the end, when I’m tired and have loads of things to do. Does that make any sense?

    • Luiz Otávio Barros
      Luiz Otávio Barros
      Posted at 20:21h, 10 fevereiro Responder

      Hi, Ricardo
      Thanks for your comment!
      I think I know what you mean.
      In an imaginary (but, oh, ever so ubiquitous!) ELT world in which classroom events can be divided neatly into “accuracy” and “fluency”, it’s relatively easy to provide corrective feedback. But students’ verbal behavior in real classrooms, especially after A2, is way too complex and unpredictable to be dealt with on a “correct during the accuracy phase” / “withhold or delay correction during the fluency phase” basis.
      True, a lot of the meaning-oriented work we do in class doesn’t sit comfortably with on-the-spot interventions. But a lot of it does. And here’s where things get complicated – and potentially taxing on the teacher, as you said. What criteria can we use to decide what gets corrected and what doesn’t?
      1. The student’s degree of introversion / extroversion / level of attainment / confidence.
      2. The degree to which students are developmentally ready to profit from that piece of corrective feedback. Is the correction anywhere near students’ zone of proximal development?
      3. The length of the faulty utterance.
      4. The likelihood of future recurrence (either within the same activity or the same lesson).
      So… When you’re drilling students or when they’re engaged in very controlled practice in pairs, on-the-spot correction is your default choice, of course. When they’re discussing the death penalty in groups, delayed correction is your default choice, too. But, as I said, a lot of what is said in class falls somewhere in between:
      “Teacher, has a test on Monday?”
      “Can anyone borrow me an eraser?”
      “Mr. Smith is married with a young woman” (comprehension question)
      “Teacher, but you said for us you will play a music today.”
      And the list goes on and on and on.
      So my contention is that we need to be better equipped to deal with students’ interlanguage all the way through the lesson and that involves thinking beyond activity / lesson phase.

  • Leo
    Posted at 21:28h, 10 fevereiro Responder

    Luis Otavio,

    I have to congratulate you on writing about something that is dear to my heart.

    The other day I was having a conversation with a good friend here in Toronto and we started addressing the thorny issue of whether to “intervene” while the students are producing the language or do delayed feedback. Unfortunately, we did not come to an agreement and he does not think intervention during a fluency stage is beneficial. I, on the other hand, beg to differ. Being trained under the CELTA and DELTA models of language teaching, I started questioning the efficacy of ‘delayed feedback’ and – just like you – don’t find it that useful in the classroom.

    I was having another chat with another teacher friend at our school and we both seemed to agree that intervention at the heat of the moment has more advantages, some of which you’ve already mentioned in the article. We even discussed the whole dichotomy “fluency x accuracy” and how we deal with feedback in these stages. I am of the opinion that there should be fluency in accuracy stages and vice-versa. We could demand more (to quote Scrivener) from our students and expect them to, for instance, read dialogues during a so-called “accuracy” task and expand on them (working on fluency). Why on earth do we only focus on accuracy during fluency stages anyway?

    What’s your take on this? I’d be very happy to hear your two cents on this matter.

    Great post and you have a fan in Canada!

    • Luiz Otávio Barros
      Luiz Otávio Barros
      Posted at 23:17h, 10 fevereiro Responder

      Hi, Leo
      Thank you for your comments – and your kind words.
      It’s funny you should mention Scrivener.
      I had his “High-Demand Teaching” session / blog / articles in mind all the way through the writing process.
      I think this sort of more lenient, laissez-faire, non-interventionist attitude to spoken errors that mainstream ELT seems to have adopted in recent years is, in many important ways, a subset of low-demand teaching, a far more complex (and far-reaching!) phenomenon, which Jim Scrivener and Adrian Underhill have sought to describe in some detail.
      Maybe low-demand teaching is a byproduct of the so-called communicative era. Maybe it’s a snapshot of the so-called digital age and its emphasis on getting stuff done / getting messages conveyed, first and foremost – preferably in 140 characters or less! Maybe it’s a byproduct of the post-Headway era, in which coursebooks are so good that the teacher may sometimes end up taking a backseat and simply going through the motions, moving from activity to activity. It’s hard to tell.
      One way or another, maybe you’ll never be able to persuade your colleague that instant correction is more effective. And maybe (s)he’ll never even meet you halfway. Who’s right? Well, both of you. After all, you’re acting according to your own sets of beliefs, which, in turn, are byproducts of your experiences as language learners, initial training, school culture and so on and so forth.
      You asked me where I stand on this issue. As I said, when it comes to feedback on form (form, forms – whatever), I tend to be a more interventionist sort of teacher. But that’s because most of the students I have taught in the last, what, 10 – 15 years, were relatively fluent, but at times painfully inaccurate. And the same applies to the lessons I’ve observed since the late 90s.
      I wonder whether I would have written this article (and, indeed, if oral correction would be something so dear to me) if I operated in a different kind of context.
      Thanks for stopping by.
      Greetings from SCORCHING São Paulo.

  • Stephen Greene
    Stephen Greene
    Posted at 22:11h, 10 fevereiro Responder

    Hi Luiz,

    I was wondering if you have noticed a difference in your attitude to correction depending on the size of the class. From my experience, the smaller the class the more on-the-spot correction I seem to do. I’m not quite sure why this should be so, but perhaps it is the more personable nature of a smaller class.

    • Luiz Otávio Barros
      Luiz Otávio Barros
      Posted at 23:28h, 10 fevereiro Responder

      Hi, Stephen
      Me too!
      And the smaller the group, the more allow myself to be sidetracked in class. And the more I allow myself to be sidetracked, the more unplugged and conversation-driven my teaching becomes. The more unplugged it becomes, the more emergent language there is. The more emergent language I have to deal with, the harder it is for me to draw a line in the sand between oral correction and linguistic scaffolding in general.
      This doesn’t happen – at least not as much and not as often – in larger groups.
      Time and time again, however, I have observed groups of 5, 6 students being taught with very very little overt correction, which makes me wonder if Jim Scrivener and Adrian Underhill are actually on to something. (see my response to Leo, please).
      Thanks for stopping by!

  • Deuseana Barbosa de Souza
    Posted at 00:34h, 11 fevereiro Responder

    I couldn’t agree more with all your ideas on this issue.

    • Luiz Otávio Barros
      Luiz Otávio Barros
      Posted at 08:25h, 11 fevereiro Responder

      Thank you, Deuseana!
      Who would’ve thought we’d be interacting again after all these years, uh?
      Um beijo.

  • Deuseana Barbosa de Souza
    Posted at 00:37h, 11 fevereiro Responder

    I couldn’t agree more with your ideas on this issue.

  • Tim Phillips
    Posted at 22:52h, 11 fevereiro Responder

    Hi Luiz,
    If we really want to intervene in a constructive way, (and I do), I think that the interruption (giving the correct version) in mid-sentence is probably the least intrusive in terms of fluency. It can be, and usually is, followed by the student repeating the correct version and continuing what they wanted to say. This best reflects the way many parents/teachers intervene when a child is learning their first language. In small groups where good teacher student relationships exist it becomes a helpful tool for students and not a reason to feel bad. Like you, I find the intervention becomes a bit misty between correction and scaffolding. I am pretty sure the observations made here about group size and the amount of this kind of intervention happening are spot on. But that doesn’t necessarily rule out such intervention in larger groups. It may need some explanatory work on the part of the teacher initially but I believe the advantages out weigh the disadvantages. Because of numerous student comments, one of my differentials these days is “He corrects my mistakes”

    • Luiz Otávio Barros
      Luiz Otávio Barros
      Posted at 00:12h, 12 fevereiro Responder

      Hi, Tim
      Thanks a lot for your post.
      You bring an interesting angle to this discussion: What if mid-sentence correction is not the boogie man in the closet after all?
      Funny you should mention parent-child correction. As a native speaker of English, that’s probably how your parents corrected you – is that right? And maybe that’s how you correct your kids – if you have kids.
      So maybe mid-sentence correction is in the “OK to do” section of your belief-system because you experienced it first-hand as a learner. And then, when you started teaching, because it made sense to you on so many levels, you found a way to make it work with your students. And maybe the success you encountered reinforced this particular belief.
      Long story short: I don’t personally believe in mid-sentence correction, but you do and you have your reasons to. This goes to show, I think, how a teacher’s sense of plausibility is much more experience/reflection-driven than it is theory-bound.
      Now, where do I stand on this?
      I’m not a native speaker of English, which means I went through a completely different learning process. In class, I used to like (love!) being corrected, but I remember quickly losing my train of thought whenever the teacher interrupted me. If it was a very mundane topic / simple sentence, I’d usually (begrudgingly!) find a way to go back to what I was saying and correct myself. But if I happened to be talking about something more complicated, the teacher’s correction would usually go in one ear and out the other – sometimes not even in.
      A few years later, as a teacher, I saw the exact same thing happening in my lessons. So, on a purely intuitive basis, I stored mid-sentence correction in the “never ever ever” section of my belief system.
      Cut to the mid-90s, when I got involved in teacher education and started paying closer attention to the teachings of people like Peter Skehan and his eye-opening research on task demands, cognitive load and language processing. In hindsight, that was just the theoretical framework I needed to understand why mid-sentence correction never used to “feel” right: The more students are struggling to convey their meanings (content), the fewer attentional resources they will have at their disposal to focus on language and internalize the incoming correction. This means that cutting students short before they’ve finished saying what they had to say might place even more cognitive demands on their shoulders – demands that less analytical / field-independent students might have trouble meeting.
      Now, having said all that…
      1. When you’re learning your native language, maybe meaning and form compete against each other for attention in a different way. Maybe there are slightly different cognitive demands at play, which means that mid-sentence correction might -and I say might – be far less intrusive. So this may – and I say may – explain how mid-sentence correction found its way into your belief system / teaching repertoire and why you can use it so successfully, like you said.
      2. In my lessons, I have often used mid-sentence correction as a sort of shock therapy. You know those very fluent students with fossilized mistakes that just won’t go away? In those cases, drawing students’ attention to the mistake in the middle of the sentence (maybe through gestures?) often helps – not least because they’ve gone through the experience of trying to fix that mistake so many times that it’s nearly as second nature as breathing.
      Sorry for the long post, but oral correction is really one of those thorny, thorny issues that we can never address by simply saying “you should do this / you shouldn’t do that.”
      Greetings from SCORCHING São Paulo.

  • Damian Williams
    Damian Williams
    Posted at 11:18h, 12 fevereiro Responder

    Great post, Luiz, thanks. I definitely agree that there’s a lot more scope for intervention. I always find this lacking when observing teachers, even those who are experienced. It’s quite a refined skill and takes a lot of effort and practice to be able to do it effectively and without placing the spotlight on the learner too much, but little and often is the key I think. Also, learners rarely complain about being corrected too much. They do complain about not being corrected enough though.

    However, I do still think delayed feedback is important (and I’m sure you do too), and not only as a way of correcting. It can be used as a springboard to teach new lexis by exploiting new language which arises incidentally (‘correct and collect’), as well as praise individual learners for good or creative use of language, which can then be shared with the class. Also, perhaps most importantly, it provides the teacher with a record/bank of common errors and incidental language over a number of classes, which can then be used formatively to help plan future classes and review/recycling activities.

    • Luiz Otávio Barros
      Luiz Otávio Barros
      Posted at 11:36h, 12 fevereiro Responder

      Absolutely, Damian. Agree 110%.
      “Delayed praising” has always worked wonders in my lessons and, not surprisingly, the forms I draw their attention to usually stick like glue.
      And your last point is particularly important: Keeping a record of students’ mistakes over a number of classes, which, unfortunately, relatively few teachers do.
      And here’s a third one:
      Assuming that corrective feedback can both (A) help students gradually eliminate the error being corrected and (B) improve their overall ability to self-monitor (because they know there’s a classroom culture where accuracy / complexity / range matter), then maybe delayed feedback, for all its short comings, is a valid way to address (B).
      Thanks for your comments! 🙂

      • Teresa Gomes de Carvalho
        Teresa Gomes de Carvalho
        Posted at 17:45h, 22 fevereiro Responder

        I also agree 100%.

        • Luiz Otávio Barros
          Luiz Otávio Barros
          Posted at 18:18h, 22 fevereiro Responder

          Hi Teresa,
          Thank YOU for taking the time to write such a thoughtful reply.
          You raise a number of very valid points and, yes, we’re on the same page.
          You talk about monolingual vs. multilingual groups. Good point. I actually think that a lot of our leniency towards errors stems from the hardly-ever questioned mantra whereby we should “focus on errors that interfere with intelligibility.” In a monolingual setting, relatively few errors hinder message conveyance.
          Thanks again.
          Um abraço!

  • Teresa Gomes de Carvalho
    Teresa Gomes de Carvalho
    Posted at 20:37h, 21 fevereiro Responder

    Hi Luiz Otávio, I agree with every word you say, especially # 4 when there’s an opportunity for retrial. And I also agree with Stephen Greene on the fact that smaller classes make on-the-spot correction more feasible. Being corrected on the spot works for me and I believe that asking learners to try again using the correct form gives them the ‘here-and-now’ perspective of the language and immediate feedback when there’s a communication breakdown.

    I also tend to agree with Damian Williams: In the language institute where I teach, teachers used to be trained to avoid intervening during tasks. So, when my school shifted this mindset a couple of years ago, many teachers felt reluctant to interrupt learners to correct them on the spot. Because I’d always done this in my previous jobs, I felt comfortable with explicit correction rather than the ‘recasting’ technique most teachers would use for fear of interrupting or embarrassing students. This is not to say that we have phased out delayed feedback. We still use it, but we shouldn’t be afraid of doing our jobs and I don’t think students resent being corrected on the spot.

    However, whichever road we take, we can’t forget to give students an opportunity to say it again until they get it right.

    • Luiz Otávio Barros
      Luiz Otávio Barros
      Posted at 11:36h, 22 fevereiro Responder

      Hi, Teresa
      Thank you for your comments.
      Isn’t it funny how school culture – staffroom culture, in particular, I think – can shape what we do in class perhaps more than any amount of training or theoretical input?
      Something else comes to mind: a lot of us fear embarrassing and upsetting students if we adopt a less unobtrusive, more overt and straight-forward approach to feedback. Maybe we’re afraid of that because we’re teachers – people who perhaps are naturally more inclined to listen carefully, respect individual differences, avoid discomfort and so on. So if we’ve become teachers, aren’t our interventions less likely to upset students anyway? Once I remember observing a DOTE lesson (that’s what the DELTA used to be called – sort of) that for whatever reason generated lots and lots and lots of mistakes (recurring, one-offs, grammar, pronunciation – you name it). There was hardly any corrective feedback at all. The teacher, who’s one of the sweetest people I’ve ever met, told me she was afraid to “put learners on the spot and embarrass them.” I don’t remember exactly how I gave her feedback way back then, but if it was today, I think I might tell her:
      My dear, you couldn’t upset / embarrass / offend your students EVEN IF YOU TRIED. It’s just not who you are.
      Long story short: Yeah, maybe we’re unnecessarily over-zealous.
      Thanks for stopping by, Teresa.

      • Teresa Gomes de Carvalho
        Teresa Gomes de Carvalho
        Posted at 17:11h, 22 fevereiro Responder

        Thanks for your prompt response, Luiz. This is a great post in that it discusses an issue that is both grounded on one’s cultural and emotional backgrounds. Regardless of the correction technique prescribed by the school, younger teachers might not feel comfortable interrupting older learners, for example — I’ve been there! As a mentor, I often give feedback to my peers and I do my best not to sound judgemental or too prescriptive when I don’t need to be. It also has to do with our own attitudes towards others. As you put it, as teachers we’re expected to “listen carefully and respect individual differences.” By that we mean “non judgemental.” But we have to accept the fact that we’re language teachers and that yes, we need to sort out what is linguistically acceptable and what is not. Once my upper-intermediate students and I were sitting in a circle and sharing our stories. It was a very enriching moment and we were all enjoying it when I corrected one of the students right on the spot and asked her to rephrase it and move on. When we finished the task, I got up and discussed some error correction techniques with them. I told them that of course I was listening for the content of their stories but I couldn’t forget that the main reason why I was sitting in that classroom was precisely to make sure that whatever message they wanted to get across would be understood by any English speaker. I’m a language teacher and this is part of my job.

        As long as we make students feel comfortable and get them to trust us, I don’t think immediate feedback as a problem. On the contrary, I see benefits mostly. We can let students know what kind of feedback they will receive during a task. Now my school has realized that delayed feedback among other teaching practices hasn’t added much to students’ fluency, so we’ve made some major changes towards accuracy and fluency.

        For many years, I used to teach business English to groups of highly demanding students who needed to sound accurate and make a good impression in meetings and travel. Therefore, on-the-spot correction seemed the right route to take. And the results were satisfactory in the sense that students were able to rephrase whatever they’d said wrong and move on. We do need to be highly selective of when we should interrupt students during an activity and choose mistakes that actually break down communication.

        Immediate feedback is particularly important in monolingual classes because students are able to make themselves understood due to the fact that they share similar error patterns. Because mistakes can go unnoticed, it is our job to signal to them that an English speaker or a speaker of another language would not understand what they’re saying. It’s a completely different story when you have Chinese students trying to get their ideas across to French and Russian students. A minor English pronunciation mistake for a Russian is likely to become totally alien to a Chinese or a French speaker.

        In short, my experience with immediate, explicit feedback has led me to believe that it should be used more often than some communicative approach advocates or schools prescribe.

  • Daniel Cartwright
    Posted at 08:27h, 09 maio Responder

    The way I think of it is this: what makes you a safer driver: being told “drive safely” or having a near miss with a truck on the highway? I feel that when students are forced to correct themselves (even mid-sentence) it gives the “truck on the highway” experience.

Post A Comment