A lot of what we do in class tends to be related to grammar in one way or another. This post, which won the British Council Blog of the Month Award in Feb 2013, examines some of the questions that we should ask ourselves when we plan a grammar lesson.
This article is based on four broad assumptions:
a. By “planning a grammar lesson”, I mean planning a balanced lesson with an overt focus on form at some point, rather than grammar in isolation, of course. So, it could be reading or listening followed by grammar and then by speaking, for example.
b. This is not an article about emergent language and how it can provide the basis for an organic, student-driven, books-closed grammar lesson, in the best Dogme tradition. In this article, I am specifically referring to those everyday grammar lessons based on the sequence pre-determined by whichever coursebook you may be using.
c. I have finally made peace with PPP. After spending most of the 1990s criticizing the presentation – practice – production “straight jacket”, flirting with task-based learning and eventually coming up with an alternative framework myself, one day I realized that I’d been waging an unnecessary war. As it turned out, all the non-PPP alternatives I’d considered had “presentation”, “practice,” and “production” elements, regardless of how they were labeled (e.g.: “enabling task” rather than controlled practice) and sequenced. In hindsight, Jim Scrivener’s ARC model seems timeless and uncannily prophetic.
d. Focus on form or focus on forms? Error or mistake? Use or usage? Learning or acquisition? Honestly, I don’t care. At least not as much as I used to. I have grown somewhat tired of semantic hairsplitting, and, in this post, I will use these terms interchangeably.
So, with a-d out of the way, here are four questions I would encourage you to ask yourself when you plan your next grammar lesson.
Question 1: How much do students already know?
A few years ago, it was relatively easy to assume that what hadn’t been formally covered in class was bound to be “new” and should, therefore, be taught from scratch. Not anymore. In ELT, the line between “old” and “new” seems to be getting hazier by the hour and I’m not sure we’ve all caught up to this new reality. Students are exposed to so much English outside the classroom that it’s probably safe to assume, for example, that on “present perfect day”, at least some of them will have a rough idea of what it is, what it means, and how it’s formed. Some might even be able to use a few formulaic chunks (e.g.: “Have you ever…”) spontaneously in communication. This means that when you plan your next grammar lesson, before the actual “presentation phase”, you might need to use some sort of quick diagnostic task to check how much students already know. Seems straightforward enough, doesn’t it? Well, it’s not.
Here’s your first dilemma: Should you try to assess how much students already know (“I ___ [live] in London for 5 years”) or how well they can use the new grammar? (“Oh, Mario, so you study French? Where? How long? Tell me more!”), which might be harder to do but ultimately more revealing in terms of students’ actual stage of development? And whichever your choice, what if you find out that, say, 30% of your students can use the “new” grammar, 30% can recognize it, and 40% don’t have a clue? Surely you need to help ALL students make sense of the structure of the day (to borrow Scott Thornbury’s term), which raises an important question: What was the point of the diagnostic task in the first place, then? Mostly to make the “new” grammar more salient, I believe. Let me explain.
By setting up diagnostic tasks designed to “trap” the new grammar, we can help students notice certain gaps in their repertoire: “Gee, I didn’t know how to express that idea – wonder what I should’ve said” or “Would have went or would have gone? Let me pay attention to the listening.” This will, in turn, create the need for the “new” grammar and, therefore, make it more immediately noticeable in whatever texts or dialogs we use during the presentation phase. Read this summary of Merill Swain’s Output Hypothesis to learn more.
Question 2: How much noticing and self-discovery will students really do?
As far as grammar “presentation” goes, the orthodoxy of the past 25 years, since the first edition of Headway came along, has been remarkably consistent:
First, students read/listen to a text or dialog and answer comprehension questions.
Next, they’re encouraged to notice three three or four examples of the “new” grammar.
Finally, they answer “discovery” questions to help them understand the new rules. This is a sound, dependable model, I think. But in the messy world of the classroom, things are nowhere near as simple, and there are times when your best bet is probably to tweak the text – noticing – analysis paradigm. Or to ignore it altogether. Here are three things to keep in mind when you plan a grammar lesson:
(a) Sentences illustrating the “new” grammar must be short, clear and memorable. If you don’t like the text/example sentences in your coursebook, and you don’t have enough time to google better alternatives, remember you can always go unplugged. Use students’ anecdotes, talk about something interesting that happened to you, tell a story – you name it. Talk to them! Then either present – yes, nothing wrong with this verb – the new patterns or highlight them more inductively via dictogloss, for example. At the end, go back to the coursebook for practice.
(b) If you use the examples and grammar discovery questions in the book (e.g.: “Is sentence A about the present or the past?”), be sure to give students enough time to answer them individually first, and then, check in pairs. If you read the questions out loud straight away yourself, your intonation might reveal the correct answers too soon, especially in yes/no questions.
(c) Be on the lookout for discovery questions that are too obvious. Here’s an example:
Sentences from the text: “I’m afraid of flying.” “I’m interested in learning languages.”
Grammar discovery question: “After prepositions, we use [ING / infinitive].”
When you plan your next grammar lesson, look at the discovery questions in your coursebook carefully – especially those dealing with form – and ask yourself if they’re challenging enough. If not, either change the questions or explain/elicit the rule yourself. Again, nothing wrong with a quick, clear, and memorable teacher-led explanation.
Question 3: What is my students’ first language?
I think it would be fair to say that most adult students – especially field-independent, analytical learners – will try to make sense of the “new” grammar by comparing it to a similar structure in their L1 or in another foreign language they might know. Most of them will attempt to map out some of the main similarities and differences between the two languages, whether or not we try to stop them. This is a long and unnecessarily contentious discussion, and it goes way beyond the scope of this article, I know.
But one thing seems clear: If you and your students speak the same language, you’re much better equipped to anticipate most of their difficulties and decide how much emphasis to place on meaning and form depending on the structure at hand. For example, if you’re teaching the so-called first conditional to a group of Portuguese speakers, you won’t need to spend more than a minute or two helping students grasp its meaning – even if your coursebook (which is probably targeted at students of ALL language backgrounds, anyway) suggests otherwise. The “first conditional” in English and Portuguese is fairly similar. So, in this particular case, it’s probably form – not meaning – that deserves more attention: If he goes or If he will go? If he go or If he goes? Pronunciation might prove tricky for Brazilians, too: If he (weak h) rather than “Ifee” he.
So when you plan your next grammar lesson, look at the grammar discovery questions in your coursebook, think about your students’ L1 and consider the extent to which the meaning/form/pronunciation balance seems right. If it doesn’t, shift the focus yourself.
Question 4: How much grammar information can I expect my students to handle in one go?
If you’re planning your next grammar lesson using a coursebook, you’ll find that it’s virtually impossible to get away from the paradigm I described in question 2: text – comprehension – noticing – analysis. The amount/depth of grammar analysis, though, still varies considerably form title to title. If you feel that your coursebook tends to go overboard and tackle far too many rules, shades of gray and subtleties at the same time, consider breaking your presentation down into two parts. Here’s how:
First ask yourself:
“What exactly do I want my students to be able to do with the “new” grammar at the end of this lesson or series of lessons?”
Think of some sort of less controlled, “real life” activity, involving interaction, negotiation of meaning, and the use of a variety of language items rather than only the “new” grammar – maybe the one proposed in the lesson itself. For example, if you’re teaching present continuous for temporary actions, maybe the final activity could involve some sort of role play in which students tell each other what they’re up to and so on. Then, ask yourself:
“What examples of the new grammar can I realistically expect them to use during this less controlled activity?”
This will help you determine how much detail really need to learn before they embark on the final speaking activity. So, in the present continuous example, consider delaying the teaching of stative verbs, for instance. Why overwhelm students with rules and exceptions when you can deal with the difference between I think vs. I’m thinking at the end of the activity, possibly on a remedial basis? The same applies to complicated spelling rules – students don’t need to learn them before the final speaking task.
I hope these questions give you some food for thought when you plan your next grammar lesson. By the way, the tree is meant to remind you that, despite this article, I have not forgotten that language learning is essentially unpredictable, non-linear, and chaotic. I just don’t think that it’s always possible to teach a pre-defined grammar syllabus in an organic, emergent, laissez-faire sort of fashion.
And trust me – I’ve tried.
Thanks for reading.
Luiz, what you said about the PPP model made me sigh in relief. I never really understood why people shunned it if what they were proposing was basically the same thing with a different name.
One other thing I have to add and which takes me back to my pre-course interview for the DELTA is: I wrote somewhere that I kind of knew a certain listening task from the coursebook wasn’t going to work (it was not challenging enough) but I used it anyway. The tutor then asked me why I did so and my answer was ‘I didn’t have time to come up with a better activity’.
This is how I very often feel about the discovery questions used in coursebooks. They are may not always be good, or focus on things that are useful for Brazilian students, but there are times when I don’t find time to change them. As sad as it might seem, the biggest advantage of the coursebook is that it’s there (as opposed to authentic material you’d have to find on your own or something you could produce yourself), and that alone makes up for a lot of its shortcomings, doesn’t it?
Hi, Ricardo
I think people have traditionally frowned upon PPP (myself included) for five reasons:
1. It bears no resemblance to how students’ interlanguage develops. Learning is organic, PPP is atomistic. Yeah, I get that.
2. Students’ performance in the final task doesn’t always mean that they’ve learned the new items. It’s often “conformity” (Jane Willis) rather than true accuracy. Of course. Mastery takes time, repeated exposures etc etc.
3. Not all classes begin with the first P, move on to the second P and so on. There should be room for all kinds of sequences: Presentation – Practice – Production two weeks from now OR Production (diagnostic) – Presentation – Practice – Production (task repetition, which is in vogue, anyway) OR Production – Presentation – Production – Practice (on a remedial basis).
4. There are Ps and Ps, of course. Does the first P HAVE to be a teacher-led presentation? Does the second P HAVE to include oral drills? No, not necessarily.
5. Most grammar lessons are interwoven with skills work of some sort. So where do texts / listening / free-speaking tasks come in? What P do they belong to? (This is why Scrivener’s ARC model has stood the test of time beautifully – a MUST-READ by the way).
So if you dismiss PPP out of hand because of reasons 1 and 2, that’s where the discussion ends. There’s no arguing that PPP does not resemble naturalistic learning processes and so on and so forth. But reasons 3 – 5 are a different story. If they make sense to you, then chances are that you’ll be organizing your grammar teaching around some sort of framework that has all the PPP elements, but not necessarily in that sequence or enacted in a certain way.
Even books like Cutting Edge (the only “task-based” title to gain mainstream acceptance, I think) have lots of PPP in the core units. The so-called “tasks” sort of run parallel to the main syllabus and do not try to present and practice new language (apart from some formulaic chunks).
As to your second point, yes, coursebooks can be extremely helpful. Luckily, modern titles are way, way better than the books we had in the 80s and 90s. What we need now, I believe, are books that are targeted at more specific demographics, especially in terms of cultural background and FIRST LANGUAGE. Why should Brazilian teachers be forced to use the same books used in China or Saudi Arabia? Luckily, this is about to change.
Tavinho, I always enjoy reading your posts. They kind of reflect the struggles I still carry out in my attempts to provide a good grammar lesson. I feel that my grammar explanations only work when I find natural usage patterns that lend themselves to practice. I guess the whole grammar thing is lost by then, but the pattern practice can still save me. I resort to translation a lot too, which can be such a helpful tool. I’m particularly annoyed to have grammar aspects that only occur in written discourse being practiced in a role play in the coursebook.
When i ask the question ‘what aspects will the students be able to use at the end?’ I feel a certain sense of defeat whenever I’m trying to go over the grammar boxes in Passages.
I’m afraid my students mostly have pattern practices when I find a logical, frequent pattern I can use myself. bjo grande
Cris,
Thank you for stopping by. I always enjoy reading your comments, too, and I hardly ever disagree with anything you say.
I think there are two issues here (let’s call them A and B) and we need to get A out of the way before we tackle B.
The first issue has to do with w h e t h e r (at all) there should be these less-controlled, quasi-communicative activities at the end of a grammar cycle. Some people would argue that, no, the “last P” means absolutely nothing in acquisitional terms (evidence of learning) so it’s no use trying to devise freer tasks to trap the new grammar in the same lesson. Macmillan’s Inside Out, for example, has chosen not include semi-controlled activities at the end of each grammar cycle – and they’re fairly explicit about that in the teacher’s manual.
But, you see, I’m not intellectually ready to dismiss these activities out of hand. Even if we assume that whatever students produce is evidence of something else rather than deep learning and even if we assume that this sort of freer use won’t necessarily speed up acquisition bla bla bla, we’re still left with the question of face validity: What if it’s important for s t u d e n t s to f e e l that each grammar cycle came to a close? What if the perception that they’re rehearsing for real life has more prima facie value than we give it credit for? Surely that helps to enhance learning somehow? So, with issue A out of the way, we’re still left with another question. What language elements lend themselves better to this sort of freer task (which is the thrust of your post, right)? I guess I could divide my answer into three parts: for sure, possibly, no way.
1. For sure: functional language and lexical phrases, especially because it’s memory rather than long-term acquisitional processes that are at stake there. In other words, if we want students, for example, to memorize and retrieve a sentence like “How long does it take to get there?”, then it makes perfect sense to ask them to put it in circulation during the final activity. If they use it correctly, true, that may not be evidence of deep learning, but of conformity (in Jane Willis’ words). But that’s what chunk learning is all about, right? At least initially.
2. Possibly: grammar – and, for the sake of argument, I’m drawing an artificially clear line here between grammar and lexical chunks. As I said, the new stuff will take a while to sink in and whatever students produce at the end of the grammar cycle won’t necessarily tell you much about their evolving interlanguages (i.e., how much learning actually took place). Having said that, though, we must remember that freer tasks will often generate the sort of pushed output that will encourage further noticing down the road. (“Gee, it was hard to say sentence X – let me pay attention to how it’s formed once more.”)
3. No way: As you rightly pointed out, spoken English should not be written English said out loud. Certain elements of grammar seem to lend themselves much better to written English, so, in that sense, the “last P” should, in many cases, be a piece of writing, I think.
Absolutely true, most professionals in the field openly declared that PPP does not fit in the Communicative view of TEFL. Some years age, I read something about lesson presentation by J.Harmer and he ,really perfected the PPP by giving it another dimension. Through his famous ESA method. I tried it and it worked better than PPP. Once again thank you for your humility.
Hi, Madadi
Thanks for stopping by.
Yes, Harmer’s ESA is a very sensible model. It’s a little different from PPP in the sense that you can go back and forth between the different “stages”, which is a good thing, of course. I also like the increased emphasis it places on engagement and the fact that “activate”, to a certain extent, encompasses both practice (depending on the kind) and production.
I’m not sure I’d call it a method, though. I like the word “model” better.