F is for Focus on Form

13 03 2011

In his absurdist, mildly funny novel Nowhere Man (Picador, 2004), Aleksandar Hemon describes a scene where the protagonist, a Bosnian, has applied for a job as an English teacher (‘strictly out of despair’) in an ESL school in Chicago. He is given a tour of the school, and visits an advanced class where there is a discussion in progress about Siamese twins:

“I must say,” the man whom I recognised as Mihalka said, “that it is not perfectly pleasant when I watch them.”

“They are monsters,” said a woman in a dark, stern suit…

“They are humans,” Mihalka said, then lifted his index finger, enunciating an important statement.  “When I had been a little child, I had had a friend who had had a big head…. Every child had told him about his big head and had kicked him with a big stick on his head.  I had been very sad,” Mihalka said, nodding, as if to show the painful recoil of the big head.

“We are learning Past Perfect,” the teacher said to us, and smiled benevolently…

“I must know Past Perfect,” Mihalka said, and shrugged resignedly, as if Past Perfect were death and he were ready for it.

The scene nicely captures a number of the tensions that characterise interaction in the ESL/EFL classroom, not least the tension between, on the one hand, meaningful interaction (“Let’s talk about Siamese twins”) and, on the other, a focus on form (“Let’s use the past perfect”).

(Normally, of course, the focus on form is engineered by the teacher, not the learner. What’s interesting, in this case, is Mihalka’s dogged – if flawed – attempts to use ‘the structure of the day’. Is this because he is conscious that the teacher’s agenda is primarily form-focussed? Or is he the kind of learner who likes to try new forms out for size? Well, we’ll never know.)

Just to remind you, a focus on form “overtly draws students’ attention to linguistic elements as they arise incidentally in lessons whose overriding focus is on meaning or communication” (Long 1991, quoted in Doughty and Williams 1998, p. 3). Typically, this might take the form of overt correction, or of gentle nudging, e.g. by asking for clarification, or by re-casting (or reformulating) what the learner has said. This incidental approach contrasts with the more traditional and deliberate approach, where teaching is based on a syllabus of graded structures (or forms), and these are pre-taught in advance of activities designed to practise them – what Long called (somewhat confusingly) a focus on formS.

A focus on formS (plural) entails the pre-selection and pre-teaching of discrete items of language (it is thus proactive), whereas a focus on form is essentially reactive, entailing “a prerequisite engagement in meaning before attention to linguistic features can expect to be effective” (Doughty and Williams, ibid. p. 3).   A focus on formS presumes a PPP methodology, where presentation of pre-selected and pre-graded items precedes production, and where it is assumed that fluency arises out of accuracy.  A focus on form, on the other hand, fits better with a task-based approach, where learning is driven solely by the need to communicate and where, as in first language acquisition, accuracy is late-acquired.

Focusing on the form of learner language that has emerged in classroom interaction is also a mainstay of the Dogme philosophy. As Luke Meddings and I point out (in Teaching Unplugged):

Focussing on learners’ lives means that the language that emerges in class will be relevant to them, but there is still work to be done if both you and they are to make the most of it. This is where a focus on form comes in (p. 60).

In our book, we offer some strategies as to how to exploit the language that emerges in classroom interaction so as to incorporate a focus on form, without sacrificing real communication. These include:

1.                  Retrieve what the learner has just said.  Otherwise it will just remain as linguistic “noise”. This might mean simple making an informal note during a speaking activity, or, at times, writing the learner’s utterance on the board.

2.                  Repeat it.  Repeat it yourself; have other learners repeat it – even drill it! Drilling something has the effect of making it stand out from all the other things that happen in a language lesson.

3.                  Recast it.  Reformulate the learners’ interlanguage productions into a more target-like form. This is not the same as correction. It is simply a way of indicating “I know what you’re trying to say; this is how I would say it”.

4.                  Report it.  Ask learners to report what they said and heard in group work. Apart from anything else, knowing that they may have to report on their group work encourages learners to pay attention to what is going on.

5.                  Recycle it. Encourage learners to use the emergent items in new contexts. This may be simply asking for an example of their own that contextualises a new item of vocabulary, or it may involve learners creating a dialogue that embeds several of the new expressions that have come up.

I’m now wondering: in the case of Mihalka, in the ‘Siamese Twin’ lesson quoted above, which of these – if any – might have been the most effective strategy?

References:

Doughty, C., & Williams, J. (eds.) 1998. Focus on form in classroom second language acquisition. Cambridge University Press.

Meddings, L., & Thornbury, S. 2009. Teaching Unplugged: Dogme in English Language Teaching. Peaslake: Delta Publishing.


Actions

Information

46 responses

13 03 2011
osnacantab

As it happens, it was also in Bosnia, in Sarajevo, that I was teaching a small group of pilots. They had decided they wanted to discuss: “My ideal woman.”

According to my notes, one of them said:

“My ideal woman is having big breasts,”
“Your ideal woman has big breasts” I dutifully recast.
“As I am say” said the pilot: ” My ideal woman is having big breasts.”

I feel apologetic, but I cannot help stubbornly wondering if even the approved 5 R approach – Retrieve, Repeat, Recast, Report, Recycle is likely to bring about any permanent change in many learner’ correct use of form.

In the first instance, it is clearly unlikely that one treatment will achieve anything. And even if a learner is able to pay attention to form in the context of a language (as opposed to a communicative) session, surely the chance of transfer to the hurly burly of real, spontaneous live communication away from the learning context is even less likely to occur.

Introspecting I can see that I am extrapolating from the fact that I, personally, do not happen to learn languages by paying attention to form. Perhaps many people do. Can anyone tell me/us if there is any substantial evidence, anecdotal or even heavier that suggests that attention to form brings permanent results in, particularly, spontaneous oral communication?

Dennis

13 03 2011
Mr Darkbloom

The first thing to say about the pilot’s cheerful utterance is that communicatively there’s very little wrong with it. The change in aspect (has/is having) is a relatively insignificant one here in terms of meaning.

I personally wouldn’t lose sleep over it, but depending on your learners needs, why not highlight the difference between yours and his versions… after the conversation? This would bring more attention to it – especially if you write the two versions on the board and discuss/elicit the difference.

Repeated occurances of these verb forms and occasional focus on them will surely go a long way to helping the learner use them – though this may take a while of course. Patience is the key here.

“I, personally, do not happen to learn languages by paying attention to form.”

I think you may be interpreting ‘form’ way too narrowly here. I’d say it’s probably impossible to learn a language without paying attention to form to some degree. I mean, how else would you know the difference between a question and a statement? Or how to express past or present states or actions?

14 03 2011
Scott Thornbury

Dennis asks: Can anyone tell me/us if there is any substantial evidence, anecdotal or even heavier that suggests that attention to form brings permanent results in, particularly, spontaneous oral communication?

Well, this question is tied up with all sorts of complex and interrelated issues, not least whether classroom instruction is more effective/more durable/quicker than naturalitsic acquisition, and, if so, whether this is because of explicit instruction (including rule statements), and whether explicit instruction is synonymous with a focus-on-form, or are there other ways of raising learners’ attention about form-function mappings that are more implicit, e.g. reformulating/recasting, input flood, and so on. Nevertheless, as Ellis (1997) puts it “there is ample evidence that the acquisition of at least some linguistic structures can be permanently influenced by instruction” (p.82).

In ‘How Languages Are Learned” (3rd edition, 2006) Lightbown and Spada report on a number of studies and conclude that “the overall results of the studies described… provide support for the hypothesis that form-focused instruction and corrective feedback within communicative and content-based second and foreign language programs can help learners improve their knowledge and use of particular grammatical features. The results also show, however, that the effects of instruction are not always long-lasting. This may be related to whether there is continued exposure to a linguistic feature in the regular classroom input after the experimental treatment ends.” (p.175).

In The Handbook of Second Language Acquisition (edited by Doughty and Long, Blackwell, 2003) Robert DeKeyser reviews recent literature on form-focused instruction, including some ‘meta-analyses’ (i.e. reviews of clusters of research papers) and concludes “a considerable amount of work suggests there is a positive role for some kind of attention to form, that is, either through the explicit teaching of grammar and explicit error correction, or at least through more direct means such as input enhancement. These literature reviews also make it clear, however, that relatively few studies have consisted of a direct comparison of implicit and explicit learning, everything else being the same” (p. 321).

The authors of one of these meta-analyses (Norris and Ortega, 2000) are more upbeat: “In a meta-analysis of 49 instructional studies, featuring a range of instructional options that included grammar explanations, input manipulations, practice or output treatments and provision of various negative feedback regimes [they found], at least as currently operationalised in treatments and measured in tests, explicit types of L2 instruction consistently result in more sizeable gains than implicit ones” (Ortega,2009, p. 75).

But as N. Ellis ( 2005) notes, “explicit instruction evidences greater effect on outcome measures that are themselves more explicit and metalinguistic in content” (p.326). That is to say, if you teach explicit grammar, and you test grammar explicitly, you will find a positive correlation between teaching and testing. The effect of form-focused instruction when tested by free production tasks showed only medium-sized effects.

So, all things considered, the Zeitgeist seems to be swinging in the direction of form-focused instruction – but – as ever – the jury is still out!

13 03 2011
osnacantab

Hi, Mr. Darkbloom. You write:

“Repeated occurances of these verb forms and occasional focus on them will surely go a long way to helping the learner use them – though this may take a while of course. ”

This what I genuinely doubt. TEFL teachers around the globe, for example, try for years to get learners to add the /s/ /z/ for the Third Person Singular Present Simple – without success and it is similar for aspect -and McDonalds( “I’m loving it”) have now queered the pitch permanently.

As for the way people like me learn/have learned languages my contention is that it is not by having our attention drawn to form but either by hearing the appropriate versions in context a massive number of times, and just picking them up, or, possibly, by deliberately practising patterns which clearly contain forms but where the focus of attention and what is consciously learned is a complete, usable utterance and not a form.

Dennis

13 03 2011
Vicky Saumell

I would like to tell you of my personal shift from focus on formS to focus on form.
In my teaching context, I have changed from pre-teaching language structures and doing PPP to engineering situations or projects where certain forms are bound to come up. Thus, when this happens, I have the students´ utterances as the basis of structural work.
It hasn´t been easy but it is definitely more meaningful for students.

Vicky

14 03 2011
Scott Thornbury

Hi Vicky – thanks for that comment. As ever, I am fascinated by what you are doing in your institution. What you describe is very much a (reactive) focus-on-form (in the context of meaning-driven communication), in contrast with a (pre-emptive) focus-on-formS, of the type “today is Tuesday so we’re doing the present perfect continuous”.

14 03 2011
Mr Darkbloom

Vicky,

Certainly sounds like you’re going a lot more for ‘point of need’. I am intrigued by this kind of thing… how much to focus on conversations and activities and just see what language comes up (don’t try and predict so much) and staging conversations and activities where you’re main aim is to get them to use particular structures.

14 03 2011
Stephen Krashen

Scott I am sure would be disappointed if I did not file a minority report.

In Explorations in Language Use and Acquisition: The Taipei Lectures (Heinemann) I reviewed all the studies available to that time on focus on form in general, and found that all were consistent with the Monitor hypothesis: In all cases claiming success for focus on form or formS, gains were very modest, subjects had studied the rule, were focused on form during the test, and had enough time to retrieve the rule. I included studies found in Doughty and Williams, which Scott cites in his post.

In addition, John Truscott has published a number of fine papers challenging the efficacy of error correction, as well as the idea that language acquisition requires paying attention to form.

14 03 2011
Scott Thornbury

I would indeed have been disappointed, Stephen!

As I point out to Dennis, above, some of the evidence in favour of a focus on form is susceptible to the kinds of criticisms you outline. However there are other studies that strike me as being more rigorous, as well as very suggestive, not least the one by Doughty and Varela (in Doughty and Williams 1998) on recasts. For those who don’t have access to it, here’s a cut and paste from a review I did of that book:

A recast is a form of feedback whereby the teacher reformulates the learner’s erroneous utterance: ideally, it should be done in such a way that the learner notices the gap between his or her utterance and the correct form, but not so obtrusively that it breaks the flow of talk. To use an example from the study:

José: I think that the worm will go under the soil.
Teacher: I think that the worm will go under the soil?
José: (no response)
Teacher: I thought that the worm would go under the soil.
José: I thought that the worm would go under the soil.

[p. 124]

The researchers were curious to see what effect such recasts would have on the language development of a group of non-native teenagers in an ESL science class in the US over a sequence of lessons in which they reported on science experiments they had conducted (hence the worms). For research purposes they targeted past tense forms, recasting only mistakes in this area. Compared to a control group where no such feedback on form was given, the students showed significant improvements in their use of the past over a six week period, an improvement that persisted for at least two months after the treatment. It is important to note that the students were not getting any form of explicit language instruction in this class, and only minimal grammar instruction outside it. It seems that the recasts were sufficient to do the trick.

(I should add at this point that the data for pre-and post-testing was taken from oral and written production tasks of the same type that they were reheasring, that is to say spoken and written science presentations – i.e. they weren’t being given written tests on grammar).

Lightbown and Spada (2006), commenting on the same study, add “[The students’] progress was assessed in terms of both increased accuracy and the presence of interlanguage forms that showed students were doing more than repeating forms they had heard” (p.172).

In my review, I made the following additional comments:

The researchers were also interested in the practical implications of this kind of form-focused intervention in what were essentially fluency tasks: would students be inhibited by the recasts, for example? They found that the optimal time for this kind of feedback was during pair or group work. When students were doing individual presentations in front of the class, however, such interventions were off-putting. They also recommend that the focus on form should be brief and immediate, and, importantly “the teacher must remember to pay attention to what the student is saying as well as to the formal realization of the message” (p. 136). These recommendations suggest a possible [teacher] training agenda: coaching trainees in the use of recasts through, for example, observation tasks and microteaching. The conventional wisdom that teachers should not intervene in fluency activities may also need re-thinking.

15 03 2011
Stephen Krashen

Please see my comments on this and other studies in Explorations. Special offer to Scott: Send me your mailing address and I’ll send you a copy of the book.

ALSO:

I was a guest on a TV show today, People Make a Difference (host = Danny Brassell). It is on youtube. About 30 minutes.

I discussed testing, poverty, whether American education is broken and of course the role of libraries and the data showing that libraries can mitigate the negative impact of poverty on reading achievement.

14 03 2011
keith

osnacantab:
I wonder if there is something else going on in your pilot anecdote! In Aviation English, a very important feature is ‘readback’, the repetition of instructions back to confirm understanding, and this often takes the form of transforming an imperative into a gerund, like so –
“Taxi to runway two six left.”
“Taxiing to runway two six left.”

Perhaps his pilot training kicked in and meant that the ‘ing’ form was hard to eradicate in his “readback”. Perhaps he was simply “confirming” …

“My ideal woman has big breasts”
“My ideal woman is having big breasts …”

Incidentally, and to get back to Scott’s original discussion, aviation readback is one rare example where a transformation drill actually does reflect real-life communication. And it’s also a context where a focus on the right or wrong form can have (and has had) very serious consequences indeed …

I think Scott has mentioned somewhere that the case of a pilot during an emergency landing is one of the few where even the most dedicated dogmeist would prefer not to wait for the language to emerge naturally.

14 03 2011
osnacantab

Scott and Keith and readers, what follows is DEFINITELY an aside that I was led to in search of something quite different because of Keith’s remarks. But we are all interested in examples of authentic, real world communication aren’t we? Compare oral comprehension repetition exercises you have experienced with this:

Dennis

14 03 2011
Carol

In answer to your question “in the case of Mihalka, in the ‘Siamese Twin’ lesson quoted above, which of these – if any – might have been the most effective strategy?”

I think I’d go with recasting it. Have Mikalka repeat the story while writing it ‘as I would say it’ on the board, seeking Mihalka’s approval that I was representing his ideas accurately as I went. Answer any questions about why I don’t use ‘had had’ as profusely as Mikalka had been inclined to. If those questions aren’t forthcoming, I would ask questions to try to extend the story into the past to provide examples of when we might use past perfect and to contrast with when we wouldn’t.

Then I’d recycle it. Ask learners to think of a story from their own experience or one they have heard elsewhere to share.

14 03 2011
Scott Thornbury

Thanks, Carol! An exemplary lesson incorporating reformulation. If this were a class of several students — as opposed, say, to a one-to-one class — how would you involve the other students in this process? Or, more to the point, how would you ensure that they didn’t feel that Mihalka had hijacked the lesson?

14 03 2011
Carol

In my imaginary classroom, it was a class of several students, and all of them sat with rapt attention as I reformulated Mihalka’s story😉

Seriously though, I don’t think the reformulation part would take long and it would be presented as a chance for us all to look more closely at the language, after having reacted to the content, and in preparation for everyone telling their own stories with more accuracy, rather than simply being a chance for Mihalka to improve what he had said.

I’d introduce the reformulation, after any conversation generated by Mihalka’s comment had run its course, by saying something like ‘Let’s use Mihalka’s story to look more closely at the language’ and then proceed along the lines of my last comment.

I’ve thought about whether I’d have the other students report what Mihalka had said, but I think I’d prefer to focus on the language quickly and move on to them telling their own stories, working with their own ideas and own words. To have them report what Mihalka had said would perhaps be to impose his story on them even more.

14 03 2011
Scott Thornbury

You passed my little test with flying colours, Carol!

All I would add is that — if this lesson has focused on reformulating Mihalka’s story — the sensitive teacher would make a point of choosing another student’s story to work on in the subsequent lesson. And so on. It’s a bit like the process writing practice of ‘conferencing’, which (as Tricia Hedge describes it in her book Writing, Oxford University Press, 2005) “is a face-to-face conversation between teacher and student (or group of students). As students work on a writing in the classroom, the teacher can sit beside an individual and talk about writing in progress, give support with organisation of ideas, and help to extend the student’s thinking about the topic, where appropriate. Conferencing encourages students to think about writing as something that can be organised and improved. It gives them an opportunity to talk about their writing and reflect on the process.” (p.131). The similarity with conferencing and teaching “at the point of need” (see the previous post, P is for Point of Need) should be obvious.

incidentally, Hedge also has a section on reformulating, which she describes as “a valuable technique which makes revision and editing an integral part of writing classes. It moves away from the narrow idea of ‘correction’, which often tends to focus on the surface features of language, e.g. spelling, punctuation. It gives students a chance to analyse and discuss the organisation of meaning as well as accuracy of expression in their own work and that of other students” (p.138).

14 03 2011
gw-tesol

The airline pilots example perhaps demonstrates a situation where re-casts don’t work well – when the student has clearly communicated an idea as part of an interesting discussion and the teachers re-cast is misread as praise (?) agreement (?). Or, as Keith suggests, the ‘readback feature of pilots language acts as interference.

If the pilots were making the same kind of error repeatedly I wonder if the following approach would help:

Move from the discussion to a CLL activity where groups agree on their ‘ideal woman’ and then record the description on tape. (a kind of reporting stage?)

Have another group listen and note the key points (at this stage, the activity is still topic/meaning focussed)

Teacher transcribes the tape recordings and students analyze these in the next lesson.

An extra step could be to have the students listen and transcribe as well. It would be interesting to see if the “-ing” errors persist in the transcription.

Students highlight all the “-ing” forms and as a class decide which ones have been used correctly.

14 03 2011
Scott Thornbury

Having students transcribe their own recorded output in an excellent way of engineering a focus on form. In an article in the ELT Journal, Paul Mennin describes how he used this technique to prepare groups of Japanese learners to make joint oral presentations:

Two weeks before the scheduled final presentation, each group of three students performed a private rehearsal, with me as the only listener. The rehearsals lasted approximately 20 minutes and were recorded. These rehearsals, like the final presentations, were given without the use of scripts, though students were allowed to use small cue cards. I asked the students to transcribe a five-minute segment, which included equal contributions from each of them. They first of all transcribed the extract ‘warts and all’, including any errors that they made. They produced a typed transcript with double spacing, and made their own corrections in red pen. When they were finished, I took the copy and indicated any corrections or improvements that they had missed. This completed the task, and the paper was returned to them one week before they were due to give the final presentation.

Mennin, P. 2003. Rehearsed oral L2 output and reactive focus on form. In ELT Journal, 57/2, 133-4.

Mennin notes that, In the final presentations, there was a noticeable improvement in a number of language features, particularly in the use of articles and prepositions, as well as in the overall organisation of the content.

14 03 2011
osnacantab

Mr. Darkbloom. Surely your question goes to the heart of the learning/acquisition matter: “…. how much to focus on conversations and activities and just see what language comes up (don’t try and predict so much) and staging conversations and activities where you’re main aim is to get them to use particular structures. ”

Dennis

14 03 2011
Rosa Coelho

One of the issues that surfaces in this discussion is the suitability of different forms of corrective feedback for type of error, context of instruction, etc.

I would argue that other forms of corrective feedback would have been more appropriate in the situation described by osnancatab. As already mentioned, we might speculate that the teacher’s recast might not have been perceived by sts as a form of corrective feedback. And even if the student interpreted the teacher’s comment as a form of correction, he might have been more interested in conveying meaning than in making his utterance more target-like. I’d say this is particularly common in contexts where teacher and sts share the same L1 and are able to communicate with few linguistic resources.

In situations such as the one described, clarification requests or other indicators of non-understanding may come in very useful. First, they may be more effective in drawing sts’ attention to form. They are also more likely to encourage the learner to self-repair, which in turn ‘requires a deeper level of processing than repetition of a teacher’s recast’. (Lyster, 2007. Learning and teaching languages through content: a counterbalanced approach).

I’m not suggesting we should overuse these types of corrective feedback at the expense of recasts, but I believe that ‘feigning incomprehension’, as Lyster puts it, may at times be a very effective way of pushing sts in their output.

15 03 2011
gw-tesol

On a purely anecdotal basis I can clearly remember how an explicit focus on form enabled me to better notice features in the spoken language around me when I arrived in Japan.

In particular I remember studying comparatives in a fairly dull drill kind of way, then found them jumping out of everyday speech for the first time.

Similarly, after studying imperative forms a large number of signs and spoken commands became “comprehensible input”.

15 03 2011
Scott Thornbury

Yes, the argument that a focus on form “primes” the learner for subsequent noticing of the feature in naturally-occurring input has been advanced by, among others, Dick Schmidt and Rod Ellis. This assumes, of course, that the learner will have opportunities for naturalistic exposure, without which the taught feature may never be “activated”.

15 03 2011
Scott Thornbury

Having referenced Ellis in my previous comment, I have since been searching for the exact citation in which he argues that a focus-on-form can have a priming function, and have finally found the TESOL Quarterly article in which – as far as I know – he first suggested this: it was in 1993, in a paper called The Structural Syllabus and Second Language Acquisition, in which he writes:

“Explicit knowledge can help learners to notice features in the input and also to notice the meanings that they realize. For example, if learners know that plural nouns have an -s, they are more likely to notice the -s on the ends of nouns they hear or read in input and also more likely to associate the -s morpheme with the meaning more than one. In a sense, then, as Terrell (1991, p. 58) suggests, explicit knowledge can function as a kind of ‘advance organiser’ that helps the learner to comprehend and segment the input and also as a ‘meaning-form focuser’ that enables the learner to establish meaning-form relationships” (p.98).

16 03 2011
Luiz Otávio

1. Before my Lancaster days, back in the mid-90s, I used to think that differentiating between a focus on form and a focus on forms was nothing more than semantic hairsplitting for people with entirely too much time on their hands. In hindsight, I know that’s not the case. Not at all.

The tension between (1) trying to teach discrete items (“the structure of the day”) and helping students proceduralize that newly acquired (!!) declarative knowledge through practice on the one hand and (2) providing opportunities for meaningful communication and teaching form reactively, as the need arises (or “working with what is there”, in the words of Donald Freeman) is perhaps the single most important dilemma facing our profession. It’s really the plausability of a (weak?) interface position that still seems to keep many of us ticking.

So, no, adding the S to “form” is not semantic hairsplitting, I don’t think.

And yet, I often find myself wondering what kind of impact this sort of discussion can actually have on what’s being done in classrooms around the globe as we speak.

If -and I say if- we assume that 70% of teachers worldwide have some sort of set syllabus to follow (+ tests), can we realistically expect them to create and foster a student-driven, organic syllabus running parallel to the book, on the basis of which they can implement less atomistic, ppp-like teaching procedures in their classes? This is not for everyone.

2. The more I read about grammar acquisition and despair at the extent to which what we do in the classroom actually flies in the face of recent SLA research, the more I wonder what the role of formulaic language is. If lexical chunks are, in fact, meant to be noticed, learned and retrieved as unanalyzed wholes, to what extent does what we know about GRAMMAR acquisition still hold true for chunk learning?

Mihalka’s sad attempts to experiment with the past perfect were probably a result of overteaching and “mis-proceduralization”. True. But what if the structure of the day, had, in fact, been lexis? What if it’d been something like “How long did it take X to Y?” “It took X Z time to Y” or “No matter how hard I try, I can’t…”.

Wouldn’t that kind of languge have lent itself better to a PPPish lesson? One of the Willises talks about PPP leading to “conformity” rather than accuracy. But when it comes to lexical chunks, isn’t conformity exactly what we’re looking for?

Thank you for another great post, Scott.

17 03 2011
Scott Thornbury

Thank you for another great comment, Luiz!

With regard to your questions:

“…can we realistically expect them to create and foster a student-driven, organic syllabus running parallel to the book?”

No, but then we know that second languages are acquired by some people in circumstances that defy logic, and are not acquired by others when the circumstances (e.g. rich exposure, interactional affordances, in-flight form focus) are ideal. So – there must be something else going on. On the one hand, there may well be a lot of incidental learning that occurs alongside a focus-on-forms approach, and, on the other, a lot of confusion generated by a diet of roughly-tuned (or totally untuned) input. Who knows?

In fact, the argument about the virtues or not of a reactive vs preemptive form-focus seem to imply that language learning is a relatively mechanistic phenomenon, where intervention X always causes effect Y. A complex systems view of things would dispute this. Minor interventions can have massive effects. Massive interventions may have little or no effect at all.

Which brings me to your second question: to what extent does what we know about GRAMMAR acquisition still hold true for chunk learning?

If we consider chunks to be single items, that are learned and deployed , like individual words, then you could say that chunk learning is just item learning, like any other vocabulary learning, and that this will occur through a mixture of intentional learning (e.g. memorizing lists) and incidental learning (e.g. extensive reading), in conjunction with some kind of active use and feedback. If, on the other hand, you consider chunks as potentially generative constructions, sort of low level grammar, then it might well do to direct attention, through some kind of PPP approach, to their generativeness.

For example, one of the most frequent strings in English is the formula ‘[preposition] the [noun] of the’ as in ‘at the end of the’, ‘in the middle of the’, ‘on the top of the’ etc. Through exposure, learners may (implicitly) pick up common exemplars of this pattern (e.g. ‘at the end of the day’) and they may even internalize its productive potential and generate novel constructions. On the other hand, drawing their attention to the pattern might not be a bad thing either. What goes for ‘small grammar’ like this might also apply to the ‘big grammar’ that we customarily teach (like the present perfect continuous or the third conditional) – that a combination of exposure to – and memorizing of – common exemplars, in combination with some kind of focusing of attention, might cater for both the holistic learner (or the holistic part of every learner) and the analytic one.

What’s this got to do with complex systems theory? Only that a complex systems view argues that maybe all (language) learning is the incremental accretion of ‘good’ exemplars, until a tipping point is reached, when the system restructures itself and the exemplars ‘yield’ their grammar. The role of instruction might be (a) to provide access to exemplars and to make them salient (e.g. through input flooding, consciousness-raising); (b) to push the learner towards the tipping point by, for example, encouraging risk-taking in conditions of optimal security (e.g. scaffolding conversation); and (c) generally to increase the volume of exposure and practice opportunities.

None of these objectives is inconsistent with a pre-set syllabus of forms – it’s just that a pre-set syllabus of forms tends to constrain opportunities rather than increase them – but only if you let it!

17 03 2011
Luiz Otávio

Scott,
Thank you very much. Really. I”ll read the second part of your answer again and again. You’ve given me a lot of food for thought.

16 03 2011
steph

“One of the Willises talks about PPP leading to “conformity” rather than accuracy. But when it comes to lexical chunks, isn’t conformity exactly what we’re looking for?”

Yes, but accuracy and conformity can just as easily come through a TTT model.

“can we realistically expect them to create and foster a student-driven, organic syllabus running parallel to the book, on the basis of which they can implement less atomistic, ppp-like teaching procedures in their classes? ”

Yes we can – we just need to change the order of activities in the book. I recently observed a teacher using the book who was teaching “might have done” “must have done”. The teacher followed the book which basically consisted of example sentences, rule formulation, controlled practice and finally a picture for “freer practice” PPP.

Simply by changing the order of activities – taking the picture activity first, a teacher can see what emerges which can include but not be limited to the “target structure” (Task)

Next the teacher could either feed in the target structure/and/or other useful structures OR use a native speaker model and ask the students to pick out the key structures. Followed by controlled practice if needed. (here the teacher could even use the exercises in the book) (Teach)

and finish up with either a repetition of the task – or simply find a similar task in one of the many course supplementary books. (Task)

Ideally of course the language would emerge from relevant and real life situations and conversations in the classroom. But the point is, we can still use course book materials and work with emergent language in a task teach task lesson shape………

16 03 2011
Luiz Otávio

Steph,
I’m not sure things are quite as simple as that.
The sort of deep-end approach you’re describing still has “the structure of the day” as the ultimate goal. True, starting from the final task might help to show students the communicative relevance of what’s being taught. Plus, this sort of pushed output might help to “prime” them to better attend to form during the subsequent input phase. (I’m specifically thinking of Swain’s output hypothesis here).

But you’re still expecting students to display a certain degree of mastery of a certain language item at some stage during your lesson (possibly via task repetition) and, in that sense, changing the order of the 3Ps, in my opinion, is NOT tantamount to implementing an organic, student-driven model, though we might argue that it comes closer.

I think an important difference between a deep-end, TTT model and a more orthodox PPP framework is the role of controlled practice. In the former, it’s often used on a remedial, “after the fact” basis, whereas in the latter the assumption is that it will turn “knowing that” into “knowing how.”

But, in my opinion, both are instrinsically at odds with the sort of reactive focus on form (singular) DOGME seems to advocate.

17 03 2011
Zahid Sheikh

Hello everyone,

One problem that I’m wrestling with is the importance given to tests in most institutions alongside my desire to have a class that isn’t about studying “grammar mcnuggets” (to borrow Scott’s term). Tests are perfectly suited and generally constructued, as far as I can see, to examine whether or not the students can spit up their mcnuggets on paper, as opposed to producing real language in real situations (something which I think most if not all of us are interested in). Does DOGME and/or TBT translate into “success” on the typical institutional written tests? Can we prepare students for tests via DOGME and/or TBT? I wish I could get away from tests and do something a bit more practical, but so far, I haven’t been able to. Luiz’s comment above reminded me about this dilemma. Thanks for the help.

I specifically mentioned DOGME and TBT because I’m not aware of a teaching movement/methodology that foregrounds using the language for the sake of the language quite like them. I’m sure that this issue has been addressed in the books/articles on DOGME and TBT, but I figured I’d ask anyway.

17 03 2011
steph

“But you’re still expecting students to display a certain degree of mastery of a certain language item at some stage during your lesson (possibly via task repetition) and, in that sense, changing the order of the 3Ps, in my opinion, is NOT tantamount to implementing an organic, student-driven model, though we might argue that it comes closer.

I think an important difference between a deep-end, TTT model and a more orthodox PPP framework is the role of controlled practice. In the former, it’s often used on a remedial, “after the fact” basis, whereas in the latter the assumption is that it will turn “knowing that” into “knowing how.”

But, in my opinion, both are instrinsically at odds with the sort of reactive focus on form (singular) DOGME seems to advocate.”

Hi Luiz – actually your first sentence holds the key. In fact, I would not expect the students to display any sort of mastery of a language item at all in the way that a PPP approach tends to. I would never expect any mastery during the 2nd task, but as long as the stages are clear and the opportunity is given then we give the chance for learning to occur. As a foreign language speaker myself, I can fully agree with the notion that production is in fact that last stage of learning, before it comes, raising awareness, exposure, recognition and understanding.

With the TTT model the key is, and this is what I tell my teachers, you can work with where-ever the student might be in their process of language acquisition. One of the biggest flaws of the PPP approach is the expectation that the students will actually be able to produce the structure at the end of 90 mins.

So it all goes back to lesson aims. For some students the aim might be simply the “raise awareness” or “have them notice” the particular structure or chunk. Or to have them “notice the gap” between their receptive and productive ability. For other students who perhaps have had the opportunity to “notice this gap” several times – through carrying out various tasks and activities, the aim would then be to “provide the opportunity for them to practise the language” For others who might be able to semi-produce these utterances the aim is to “provide opportunities for further practice”

Clearly the working with emergent language through meaningful conversation driven teaching (dogme) would be difficult to implement in a “pure” way when teachers have the constraints of following a text book or getting students through an exam.

As your previous post mentioned the restrictions of these kind of classes I just wanted to share with you that in fact, dogme moments, a dogme flavour combined with a task driven approach can be easily adapted to most course books with a bit of re-shuffling.

TT actually isn’t quite the same as a deep-end TBL model. A deep end TBL model focuses entirely on meaningful tasks, task cycles and language focus really is an after-thought.

A TTT model can be use to weave in explicit language items such as chunks or formulaic expressions as well as “satisfying” certain people’s need for the inclusion of “Grammar Mc Nuggets” and the need to be seen to follow the book.

So – coming back to the TTT model – the “teach” part in the middle, can be as detailed, long or short as is needed. In the teach part we can stick in the Mcnuggets but also use it as an opportunity to weave in different expressions.

The advantage is that in a mixed ability group (and aren’t they all) we really can work with the emergent language from each student. As we listen in on the students carrying out the task or discussion which we’ve made sure is really meaning based as perhaps we did a pre-task warmer, gave an anecdote etc to pull the students in – we can then write down how far each is able to go in their productive language.

So for example, some may have perfect control over “might have done” but are not using, “it could possibly have been” others may be unable to produce might have done.

So much of the “language selection work” is being done by the teacher on the spot. The teacher would then limit themselves to a few language items to then board up and work with in ways that are relevant to the class.

I encourage teachers to do the tasks from the coursebooks themselves and actually write down what kind of language they really use to complete the task. If the “book target structure” simply doesn’t work at all or sounds completely unrealistic for the task at hand – then they dump the target structure and work with the real language that they themselves used to complete the task.

You can still follow a book and do this. For exam classes, certainly the main structures needed at say CAE level are modals, correct use of relative pronouns, and gerund/infinitive…..These can all be looked at through communicative tasks and often the book at least goes some way in providing a starting point for the type of task you could use.

Many of these structures can be looked at in the context of what the students actually produce in written work or activities such as Dictogloss. Again – going back to the idea of noticing a gap between receptive and productive knowledge – and balancing the way we fill that gap with what is authentic and what “the examining board needs for a particular paper”

18 03 2011
Mr Darkbloom

Steph,

The issue of exams and pre-set curriculum is obviously a major factor in what most teachers do in the classroom. And, of course you’re right, there’s no reason why dogme moments can’t be put into a coursebook-driven lesson. In fact, I think any serious teacher has a duty to.

The thing is though, I have to say I agree with Scott and Luis that if we want to focus on the students’ emergent language, following coursebooks is just going to get in the way. Even if just for the fact that most learners seem to be more engaged with material they’ve chosen or made.

Personally, I think the best way to use a coursebook is to let the learners decide what they want to study in it – in consultation with the teacher of course and in collaboration with their classmates. If you haven’t already, maybe you could go down this road.

17 03 2011
osnacantab

Scott, you write: “….. a complex systems view argues that maybe all (language) learning is the incremental accretion of ‘good’ exemplars, until a tipping point is reached, when the system restructures itself and the exemplars ‘yield’ their grammar. The role of instruction might be (a) to provide access to exemplars and to make them salient (e.g. through input flooding, consciousness-raising); (b) to push the learner towards the tipping point by, for example, encouraging risk-taking in conditions of optimal security (e.g. scaffolding conversation); and (c) generally to increase the volume of exposure and practice opportunities. ”

I may very well be out of my depth here, but if the complex system view posits that language learning occurs at the tipping point after the incremental accretion of ‘good’ exemplars – I get the impression that many increments are required and “accretion” suggests that a long period of time is involved – isn’t it hopeful thinking, at best, a teacherly gut feeling that one has to do something, that flooding, consciousness-raising and risk taking in conditions of optimal security will help. Don’t such suggested interventions, indeed, go against a complex systems view? (c) … generally to increase the volume of exposure and practice opportunities. ” would seem to be a way forward in accordance with a complex systems view and a procedure which probably represents how language learning takes place outside the classroom.

17 03 2011
Scott Thornbury

Dennis, this question of intervention was one I desperately wanted to raise today in a session that Diane Larsen-Freeman was giving (here at the TESOL convention in New Orleans) on the applications of complexity theory to classroom teaching. She has written that there may be a case — in language teaching — for intervening at the point where what is called a ‘phase shift’ (i.e. the tipping point) is about to be reached. I suppose. if there is an analogy, it is with other complex systems, like the weather, where at certain points a well judged intervention, such as cloud-seeding, can precipitate a downfall. Certainly, other scholars who have written about complex systems in relation to language learning, such as Nick Ellis, suggest that — without some kind of intervention — the system simply stabilises: what we used to call fossilisation. This intervention may simply take the form of highlighting regularities in the input, but it may be more interventionist still, like setting up activities which encourage the production of these patterns in some kind of meaningful, interactive context – instructional moves that are compatible with both a traditional PPP approach, or a more reactive, task-based or dogme one.

18 03 2011
steph

Dear Mr Darkbloom you said;

The thing is though, I have to say I agree with Scott and Luis that if we want to focus on the students’ emergent language, following coursebooks is just going to get in the way. Even if just for the fact that most learners seem to be more engaged with material they’ve chosen or made.

Personally, I think the best way to use a coursebook is to let the learners decide what they want to study in it – in consultation with the teacher of course and in collaboration with their classmates. If you haven’t already, maybe you could go down this road.”

I also agree that following coursebooks is going to get in the way. I’m not arguing against that at all. I’m not pitching myself against Scott or Luis. What I was saying was if people are in a situation where they have to use a coursebook (and many teachers find themselves in such a situation) then there are relatively simply measures teachers can take to increase the opportunity for working with emergent language.

True – you could say that the emergent language might not be purely “conversation driven” as the teacher is manipulating that with the TTT approach. But my point was that you CAN work with emergent language during the this type of lesson. You can also ‘hit’ most of the language tested in exams. Absolutely it wouldn’t be pure dogme as it wouldn’t be “materials light”.

Letting the students choose what they want to study could be excellent in some contexts. In other contexts less so. In Switzerland where I teach, occasionally this approach might work. On the whole, however, it’s difficult enough to get students to accept a TTT approach (they do once they see it benefits them).

So it depends on the teaching context. If I walked into the multi-national pharma companies, banks or law firms and said, OK here’s the book for the exam – now in groups flick through and tell me what you want to study, I can safely say that next month the company will have found a different language provider.

It would be “seen as” unprofessional, incompetent and these people with 90 mins a week between meetings, just want to get down to it. They would also say, “But how do we know what we need for the exam” As 80% of these courses are funded by companies who want a certificate at the end of it.

I’d also run into trouble taking the “you choose” approach with the teens at the international school. Mainly the trouble would come from the school head and the parents. But it would also arise because, like it or not, these classes are completely exam-driven and courses need to be carefully planned in order to cover the syllabus. This is hard enough for a teacher to do!

Unplugging teaching is my greatest passion…..I’m working in education where there are many external restraints to doing this to the extent of throwing away the book and/or letting the students design their syllabus. Much as I would love to do this!

So I take the approach of slowly slowly, touchy monkey. Which is what I explained in previous posts.

I’ll be running an “unplugged in principle” CELTA course this summer……where working with a coursebook will be a specific input session. If you don’t know how to do that in Switzerland – you’ll struggle to find work.

20 03 2011
Mr Darkbloom

“If I walked into the multi-national pharma companies, banks or law firms and said, OK here’s the book for the exam – now in groups flick through and tell me what you want to study […]

Stef,

I think the key here (as you touched on) is making learners understand WHY you would like to conduct the classes in a certain way. Letting them in on your knowledge about learning/teaching processes enough for them to know, not only are you professional, but you are a thoughtful, caring teacher.

I mean, in the companies, if you do indeed just go in a say ‘flick through the book’, there is a good chance it won’t work. No, what would be more sensible would be to explain the logic behind your thinking, build up a rapport with the learners, obtain as much feedback as you can and always, always be prepared to change your approach if they show some resistance. This kind of thing need not be context specific.

Having a meaningful, respectful dialogue ABOUT what can be done in the classroom is half the battle.

“I’d also run into trouble taking the “you choose” approach with the teens at the international school. Mainly the trouble would come from the school head and the parents. But it would also arise because, like it or not, these classes are completely exam-driven and courses need to be carefully planned in order to cover the syllabus.”

No one really wins in that situation. A shame.

19 03 2011
Luiz Otávio

Steph,
The sort of “semi-organic” syllabus you’re proposing can exist, of course. Any coursebook can potentially lend itself to TTT or any sort of teaching framework that bears a little more resemblance to naturalistic learning processes and is, therefore, less at odds with what SLA research has to say about language acquisition.

Whether this shift will generate measurable and long-lasting learning gains and will be perceived by the students in question as “good teaching” is still debatable, I think. But I digress.

Anyway, in my first post when I asked (rhetorically) whether we can “realistically” expect teachers to “dogme up” their coursebook-bound lessons, I kept wondering if the average teacher with two or three years experience and a CELTA under her belt has the necessary resourcefulness to:

1- allow herself to be sidetracked by the students’ agenda; 2 – be sensitive to students’ ZPD and spot all the “learnable moments” and the “oh, my God, he’s ready to internalize that” tipping points as students struggle to convey the meanings they want to convey; 3 – be prepared to engineer the students’ output in such a way that the need for “the structure of the day” arises naturally.

These are, I believe, fairly complex teaching skills since they shift the focus away from “I am teaching you” towards “I am helping you learn.”

Of all the CELTAs I’ve taught in recent years, I can think of maybe two or three teachers who were at the right level of professional development to pull off the kind of framework you’re proposing, which requires something I find extremely difficult to do well:

Intervention through observation.

19 03 2011
Dennis Newson

Luiz,

Wondering whether it is possible to expect:

“…….teachers to “dogme up” their coursebook-bound lessons….”

you give a brief sketch of the art of scaffolding learning a la Dogme, The teacher must:

” 1- allow herself to be sidetracked by the students’ agenda; 2 – be sensitive to students’ ZPD and spot all the “learnable moments” and the “oh, my God, he’s ready to internalize that” tipping points as students struggle to convey the meanings they want to convey; 3 – be prepared to engineer the students’ output in such a way that the need for “the structure of the day” arises naturally. ”

Three points.

(1) Like any other kind of good (effective, enjoyable, productive) teaching Dogme scaffolding is surely, in the last analysis, an art and an individual teacher either has the necessary charisma or they haven’t. How far can one learn to teach? Or, put differently: What kind of person can be taught to teach?

(2) On the Dogme list, over the years, two frequent questions are: (A) Isn’t Dogme teaching for experienced native speakers only? (B) Can you use a Dogme approach for beginners? These questions are not central to the present discussion, but they are, perhaps, lurking in the background.

(3) Although I’m quite certain that all contributors to this blog are concerned about the day-to-day realities of coalface (chalkface) teaching – learner expectations, the requirements of examination systems and the demands of all those higher up in the educational hierarchy, we are, surely, here at least, searchers after the ultimate truth – our discourse is luxuriously orientated towards uncovering and describing and sharing the ideal.

19 03 2011
Declan Cooley

The lesson from “Nowhere Man”:

There is indeed something ‘absurdist and mildly funny’ about the classroom scene in which the entire lesson is going awry, as evidenced by the languaging of the students, all while the teacher ‘smiles benevolently’. Bear in mind that the students may indeed have been at their tipping point for the “blossoming” of this tense [in my experience, a sign that students are at a a ‘phase shift’ point for past perfect in narratives, is when they sometimes start using present perfect as an intermediary make-shift tense for the purpose (pun only half-intended)].

The sorry state of affairs described in the novel has probably arisen as a result of some or all of the following factors:
1. The tense has been presented without contextualization of either texts or other narrative tenses.
2. The students have not been concept checked (or used a SCOBA, Lantolf,Thorne 2006) or interacted with a timeline to assist in their understanding of the usage of the tense.
3. The teacher has spent an inordinate time on form, which seems to be error-free.
4. Pronunciation has been neglected due to lack of any contractions.

To make things even more absurd, the teacher has also chosen a bizarre context of use – a discussion is rather unlikely to produce narratives; a story-telling task might be more conducive to the use of narrative tenses; there is a lot of artistic license going on here one hopes.

Nonetheless, even given all of the above factors, the lesson is still salvageable in the sense that students might be steered away from their over-use of the tense (a common occurrence if a tense has been recently introduced).

There are two courses of action:
a) immediate intervention
b) delayed intervention

For a), the teacher might
– sotto voce say ““When I was a little child,..” on first sight of an error while totally allowing the student to either continue without self-correcting or rewind a self-correct and continue (such pseudo-sub-vocal interventions can work surprising well by communicating with students on another wavelength as it were)
– stop the student with a question: the teacher can say “ ‘had had a friend’ – before what happened ? “ The student might then begin to wonder why there has been a communication break-down, which could lead him to self-repair, or at least shake loose his certainty about the usage of the tense.

for b) delayed intervention, any errors could be boarded and pair/peer-corrected later on.

19 03 2011
Declan Cooley

Or as mentioned above, b) delayed intevention could take the form of
-recording stories and transcribing, something I did for my DELTA and rewards all the work it entails
-getting ss to write their stories: this again give the option of either immediate or delayed intervention on the part of peers or the teacher; this would probably be a better platform for ZPD expansion and scaffolding than the public oratory seen in the novel.

[Excuse the rather didactic tone: writing this helps me summarise my responses to the responses; plus it is hard to get out of teacher trainer mode!]

19 03 2011
Declan Cooley

on PPP:

PPP lesson shapes seem to be everyone’s favorite whipping boy/straw man. PPP lessons and their outcomes are like three-course meals: if poor ingredients are combined haphazardly, then yes, it can easily turn into a dog’s dinner – one that a dog might even turn its nose up at. (I hope this re-hash below has some rather more delectable morsels.)

Presentation is really just a collection of techniques that can be combined and gathered into a compartmentalized stage(s) before students use language, or distributed judiciously throughout a lesson as the teacher is scaffolding the students’ language use, or used after communicative opportunities.

Practice tasks, yes, are often of the deadening gap-fill type – there are plenty of other types of practice that, when finely-tuned in terms of their design and sequencing, can work as contingent and graduated activities when coupled with teacher and paired interaction, which include peer correction etc. Done well, this indeed can produce many moments of kinda-know-how (not just know-about), especially on the form level but also by knowing-how the language might mean in various contexts.

This can totally lead to a stage where students “produce” the language without much help, but while still being internally scaffolded by recently seen rules and exemplars, externally assisted with notes or memory (from rehearsal) and/or given time to process “in slo-mo” [preparation time] before bringing rehearsal in pairs or groups to a more public open class reporting section that may bring pressure to be more accurate, all while being driven by a strong communicative purpose. I presume fewer and fewer TEFL professionals seriously espouse the view that this means that they now have totally restructured their interlanguage to the degree that use of the target language will now be flawless, in completely unsupported new contexts, at the drop of a hat. However, the production stage can give students a resultative motivating sense of having used language that was beyond (or on the edge of) their ken in a way that was enjoyable and personally communicative.
The production stage(s) may also act as a striking model by which they can compare future performance and thus notice-the-gap between their previously much-scaffolded and co-constructed output (sorry for the mixed metaphors) and what they can achieve now with very little support. The fact that they did it before can be a powerful spur to “I can do it again” and lead them to push their language to this internalized model’s level.

We tend to underestimate the complexity of what can happen in a classroom framed by an ostensibly simple PPP shape.

And can we PLEASE separate PPP from
-lockstep syllabuses
– blind structure-of-the-day approaches to course planning and lesson preparation as well as
– the tired and corpus-insensitive choice of target language items (such as the afore-blogged “backshift”)
or the idea that PPP is the only lesson shape (TTT etc)
or that PPP (or its subsidiary techniques, more to the point) can’t be used to tackle emergent language whether to recast or recycle or to highlight non-salient features
or that these are the only techniques which can be effective
or the idea that grammar is so amazingly vitally important in the larger scheme of things [lexis, collocation, priming based on genre and associations, text cohesion, and SKILLS development (!)].

19 03 2011
Luiz Otávio

Declan,
My two cents on some of your comments:

“The sorry state of affairs described in the novel has probably arisen as a result of some or all of the following factors:
1. The tense has been presented without contextualization of either texts or other narrative tenses.
2. The students have not been concept checked (or used a SCOBA, Lantolf,Thorne 2006) or interacted with a timeline to assist in their understanding of the usage of the tense.
3. The teacher has spent an inordinate time on form, which seems to be error-free.
4. Pronunciation has been neglected due to lack of any contractions.”

I think students might have ended up overusing the past perfect EVEN if the teacher had done all of the above. I don’t think we can really assume that a technically flawless presentation will necessarily enable them to deploy the target items accurately and appropriately.

“for b) delayed intervention, any errors could be boarded and pair/peer-corrected later on.”

I think this is the easy way out, of course, but not always the most effective, unless students are given an opportunity for retrial, possibly via task repetition. Delayed correction is often helpful in terms of “knowing that”, but we often neglect the “knowing how” aspect of the whole thing by not giving students an opportunity for retrial.

“…there are plenty of other types of practice (other than gap fill) that, when finely-tuned in terms of their design and sequencing, can work as contingent and graduated activities.”

I agree. If we subscribe to the idea that practice can proceduralize declarative knowledge, then it makes sense to provide practice activities that bear some resemblance to “the real thing”.

“The production stage(s) may also act as a striking model by which they can compare future performance and thus notice-the-gap between their previously much-scaffolded…”

I’m not sure all students have the necessary self-awareness (and monitoring skills) to notice the gap between what they were able to produce with more and less scaffolding. I personally find it easier to help them “notice the gap” by having them compare their production with some sort of “external” model, often via dictogloss etc.

19 03 2011
Mr Darkbloom

“PPP lessons and their outcomes are like three-course meals: if poor ingredients are combined haphazardly, then yes, it can easily turn into a dog’s dinner”

I suppose (only suppose) the three-course meal analogy would be great if we forget that probably most of the time, people don’t have three courses. What about a main dish and a drink? Optional starter/desert?🙂

When discussing this in Scott’s recent PPP post, I was careful to mention, of course, in the hands of a skillful teacher, PPP could easily be knocked into a reasonable lesson. The teacher can have lots of ‘dogme moments’, focusing on emergent language in a PPP lesson.

The main point here is that if a teacher steps into the classroom intending to consciously do a PPP lesson, they are perhaps trying too hard to control what learning takes place. Trying too hard to predict the outcome. We all know that lessons take on their own life given half a chance, going in unpredictable directions. All the teacher has to do is give assistence and direction and not stop the flow, as it were.
The teacher is crucially ready to exploit what the learners feel like expressing on any given day, rather than thinking… ‘shit, only 10 minutes and I haven’t got to the production stage yet!’

And before anybody pokes me in the ribs… No, I’m not saying every lesson should be like this (in Utopia maybe) and I’m not talking about exam preparation either.

“Presentation […] can be combined and gathered into a compartmentalized stage(s) before students use language, or distributed judiciously throughout a lesson as the teacher is scaffolding the students’ language use, or used after communicative opportunities.”

If we choose to ‘present’ throughout a lesson, I would argue we are not doing PPP as it is usually practiced anymore. Are we?

19 03 2011
Steph

“Of all the CELTAs I’ve taught in recent years, I can think of maybe two or three teachers who were at the right level of professional development to pull off the kind of framework you’re proposing, which requires something I find extremely difficult to do well:

Intervention through observation.”

Hi Luiz – do try and come to IATEFL this year if possible. Luke and Scott are holding a “Dogme symposium” on the Monday morning. Anthony Gaughan should be there too. Last year Anthony and Izzy gave an inspirational presentation on how they’ve unplugged their CELTA course in Hamburg. I interviewed Anthony, Izzy and Scott afterwards for our ETAS journal. If you want a copy I could try and organise that one is posted to you, but I’ll need an address. My e-mail is nowordssteph@yahoo.com.

Anthony’s a CELTA trainer and assessor and has basically had extremely positive experiences with unplugging his CELTA course over the last few years. In fact, as he mentioned in the interview, the grades on his courses have actually increased.

I’ve been bringing in “intervention through observation” in as many of my in-house and external workshops as possible, and people do get it – even relatively in-experienced teachers and teachers without a CELTA. There are a growing number of CELTA trainers out there who are extremely interested in this.

19 03 2011
Luiz Otávio

Steph,
What a tempting offer!

24 03 2011
Scott Thornbury

As a footnote to this discussion, I should mention that Stephen Krashen has kindly sent me a copy of his 2003 book Explorations in Language Acquisition and Use (Portsmouth, NH: Heinemann), in which he critiques the argument that a focus on form is beneficial. Specifically, he revisits many of the studies that were the subject of the Norris and Ortega (2000) meta-analysis, which argued for positive effects for a focus on form. He demonstrates (fairly convincingly, I might add) that all that these studies succeeded in doing was to demonstrate that, if you teach grammar and then test it with grammatical tests, you get short-term benefits. “Predictably, they show that more focus on form and more information presented about rules results in more conscious learnng” (p. 41).

However, the benefits that are demonstrated over the long term or in so-called “free-response tests”, i.e. free production tasks, are nugatory to say the least, and hardly merit the time that was put in, in terms of instruction. “These results only confirm that direct teaching results in a fragile kind of knowledge that is applicable only when very severe conditions are met” (p.44).

My feeling remains that, while lengthy grammatical presentations are probably a waste of time, given the meagre benefits they reap, there is no harm in — in the context of authentic language use — gently reminding learners of target forms, if it takes little or no time away from that authentic language use. There is nothing to be lost, even if there is little to be gained. Chances are, though, that some students will notice some interventions, and take them on board — much in the way that Mihalka overgeneralised the use of the past perfect — and that these ‘noticings’ will eventually establish themselves as permanent features in their linguistic landscape.

10 10 2011
Karen Bruntt

Dear Scott Thornbury

I find the excerpt from Nowhere Man very amusing and I was wondering if you know of other novels/short stories in which the teaching of grammar is included?

Kind regards, Karen Bruntt

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s