Complicating the coursebook debate part 2: can’t we just be friends?

This week I thought I would take a break from the grammar series (to be continued!) and pick up on the discussion of coursebooks that Hugh started some time ago with the somewhat optimistically titled Complicating the coursebook debate part 1. This was almost two years ago now and we never even managed to move on to part 2 (let alone 3 or 4)! In the intervening time, the debate and the general abuse of coursebooks and their writers has rumbled on with the most vociferously anti-coursebook people almost getting to the point of saying that you simply cannot learn anything with a coursebook – and what is actually ‘learned’ is done so in spite of the coursebook, rather than because of it!

Coursebooks and their apparent failure to follow Second Language Acquisition theory

The most vocal critics claim that coursebooks cannot work as they are based on three supposedly false assumptions. What follows is taken verbatim from perhaps the angriest of the anti-coursebook bloggers, Geoff Jordan.

False Assumption 1: Declarative knowledge converts to procedural knowledge through classroom practice. Jordan argues that it does not. Knowing that the past tense of has is had doesn’t mean that with a bit of classroom practice you can use had fluently and correctly in real-time communication. L2 learning isn’t like studying Geography or Biology, where declarative knowledge is primary, it’s more like learning to swim.

False Assumption 2: SLA is a process of mastering and accumulating structural items one by one. It is not, Jordan notes. All items are inextricably interrelated, and there’s no evidence that items are learned one at a time, or that any item is ever learned linearly. As Long (2015) says: “The assumption that learners can move from zero knowledge to mastery of negation, the present tense, subject-verb agreement, conditionals, relative clauses, or whatever, one at a time, and move on to the next item in the list, is a fantasy”.

False Assumption 3: Learners learn what they’re taught when they’re taught it. They don’t. L2 learners follow their own developmental route, which involves a series of interlocking linguistic systems called “interlanguages”. The route is not affected by L1, learning context, or teaching method. Teaching affects the rate, but not the route of SLA.

Let’s talk about teaching – not materials

In taking things forward, it might be better to see these ‘assumptions’ that Jordan claims coursebooks are rooted in as applying instead to teaching. I actually basically agree that all three assumptions as they are presented above are false, though I’d add a couple of caveats. However, we might also see some false assumptions underlying the declaration that they are only false for coursebook users, as in fact they can equally apply to a task-based or Dogme-influenced teacher. So let’s take them one at a time.

Passing on knowledge is a starting point, not a final destination

1    There is no doubt that “knowing that the past tense of has is had doesn’t mean that with a bit of classroom practice you can use had fluently and correctly in real-time communication.” Of course the job is not finished when irregular past tenses are presented and practised in class. But that’s true of all types of classroom teaching. Would the job be finished if the past tense of had was required during a task and ‘emerged’ or was ‘noticed’ or ‘recast’ or whatever?  Of course not! Students will continue to make mistakes because whether this information is passed on to students through a task or a ‘bit of classroom practice’, the information still remains essentially declarative at that point. By definition, declarative is something we know, while procedural is something we use.

Bad swimmer

Jordan’s analogy with swimming is surprisingly helpful here, though perhaps not for reasons he realises. During most swimming lessons above absolute beginner levels, nearly all instructors work outside the water! They are telling students what to do (declarative knowledge) and getting students to proceduralise this – often through rather synthetic (and to my mind rather monotonous) tasks such as ‘practise breathing out underwater’ or ‘swim with your legs only’. This is stuff that needs to be pointed out for learners to improve, but as a rather poor swimmer myself, not always to great effect, as I haven’t been able to turn my ‘declarative’ knowledge into action – perhaps because I haven’t sufficiently practised swimming only with my legs!

The lesson for teachers and learners is that you have to make use of the language you have ‘learned’ – and do so repeatedly over time. Certainly, a coursebook should give opportunities to make use of language and should ensure proper recycling over time, but it may not. A task-based syllabus should certainly provide opportunities to make use of language, but it may not be the same language over time. Although, of course, it might be! It all depends.

When do you move on to a new ‘teaching’?

2    This is an interesting question because while I basically agree that SLA is not a process of mastering, one by one, accumulated structural items as in some kind of building block process, you could argue that the next point that ‘Teaching affects the rate, but not the route of SLA’ is slightly contradictory. It seems clear that language learners move from more or less ungrammaticalised words to grammaring the words they know in progressively more complex ways: this is the route. So in the case of questions, students at the lowest levels will generally start by just using a word, maybe with intonation or gesture – coffee?; then move to a string of words – you want coffee?; to grammaticalised strings – Do you want a coffee?; to more complex sentences – Are you sure you want a coffee? If you were a mad person and did these as consecutive lessons, your students would not be producing all these different question types. However, even here there is an important caveat. Students can sometimes learn a piece of grammar as a vocabulary item. So a student might actually master How are you? without having a general control over inversion in questions. In his research on acquisition order, Pienemann specifically discounted these instances (See Is Language Teachable? Psycholinguistic Experiments and Hypotheses,  p58 Vol 10, Applied Linguistics, OUP 1989, and his comments on Teresa and rote learning).

Who’s assuming mastery?

With regard to coursebooks and their users, coursebooks clearly don’t actually expect mastery of their grammar ‘nuggets’ as can be seen by the fact that these tend to get repeated at different levels. Coursebooks may also repeat points within a course – or not – and they may provide speaking/writing opportunities where students could re-visit grammar – or they may not. However, the same is also obviously true of task-based or conversation-driven lessons. The conversation happens and then there is a stage where the teacher chooses to focus on particular language items. What will they be? How does a teacher decide the grammar or lexical focus that the students are ‘ready’ for? How do they decide who is ready – out of a class of 10 or 15 or 30 or more? What do the students do when they are focusing on form? Do they use materials at this point? What are those materials?

Coursebook writers are teachers

It is quite possible that the Dogme or task-based teacher chooses to pick up on the same kinds of grammar points as a coursebook, perhaps bringing in decontextualised examples disconnected to the original task (such as a photocopied page of best-selling grammar book). And given that teachers often share with writers the same concept of level with regards to grammar, then it’s quite likely a non-coursebook using teacher may focus on the same grammatical patterns  as a coursebook does. Or they may not!

And finally, how many times does the Dogme / TBL teacher focus on the same point of language before moving on? Once, twice, three times? When do you decide if something has been learned? This is just as much a problem for a ‘materials-free’ teacher as it is for those who follow a coursebook, though I think, as we have pointed out in other blog posts, there is a certain restriction in the items and staging of grammar items as presented by most  coursebooks and potentially greater flexibility that comes from interaction with students (see next post).

It’s good for all to remember we can’t really control learning

3     This is also true up to a point, but it does depend what you mean by learnt and taught. If I teach someone the meaning of a word they did not know – say, table – the student clearly can ‘learn’ it and even use it in some very limited way (for example point and say ‘table’) at that moment. That obviously does not mean it has been learnt permanently or completely:

  • It may well be forgotten if not used over time (or consciously studied and recalled);
  • Students may use it wrongly, applying it to a ‘desk’ or a ‘sideboard’ or saying it as ‘tarble’;
  • they may use it with the wrong article or no article;
  • they may not know how it is used in more metaphorical sense – e.g. lay your cards on the table.

However, all of this is true of forms which are taught or highlighted or noticed in response to tasks or conversation. Pre-selected materials, such as coursebooks or things teachers plan to teach, may carefully ensure recycling of language items (or they may not!). However, it’s quite possible that ‘teaching’ that occurs in response to a breakdown of communication will not ever be recycled! And just because you did your teaching to fill in a gap in a student’s knowledge doesn’t mean it will be directly learned – forever. It won’t.

Routes and rates – really, what’s the difference?

Then what about this idea that “the route is not affected by L1, learning context, or teaching method. Teaching affects the rate but not the route of SLA”? Firstly, this rather suggests that there is a single route (not unlike a coursebook syllabus if taken literally). However, there are clearly many routes going on at the same time – particularly when it comes to vocabulary. How many routes are there? How many can be pursued at the same time? More to the point, what on earth is wrong with speeding up the rate of acquisition? In effect, this stuff about routes and rates is a difference without a distinction. Say I teach a relative clause or chunk to a student who isn’t ‘ready’. They don’t ‘learn’ it at that point, but apparently it may lead to them acquiring it more quickly at some future time!  This is like saying someone can’t use the motorway to get to a destination – they can only use A roads. However, we can also give the person a Sat Nav or access to Google maps rather than just saying ‘go North’. That seems a good reason to teach, whether it’s with a coursebook or not.

False assumption on top of false assumptions

Which leaves two further assumptions made when pointing out these supposedly false assumptions. Dismissing coursebooks by claiming they’re based on these false assumptions also makes assumptions – and false ones at that. It assumes that the coursebook user follows the coursebook to the letter and that a materials bank that a non-coursebook user brings to a class is somehow more principled – simply by not being a coursebook. Both of these assumptions, I would suggest, are false and I’ll look at these in my next two posts on the theme: using a coursebook and principles and the  materials bank – well, within the next two years, anyway!.

All this is not to say that everyone should be forced to use coursebooks or that not using one essentially defines you as a better teacher. The points on SLA here are good ones; it’s just that we should all think about them whenever we teach. There are clearly good principled ways to use coursebooks and to not use them – and, of course, there are bad ways. And lest both parties forget, learners become fluent speakers by living the language outside the classroom. Ultimately, teachers are just bit parts, whatever our methodology.

Want to think more about your teaching? Take a summer course with us in London.

Find out more about our most recent series of coursebooks, Outcomes.

  • Matt Byrski

    Some time ago, I taught a course without a coursebook. I collected topics while doing needs analysis and based lessons on them. It was hard work and lots of preparation. Some lessons were better than others. It was a challenge for me as I am nowhere near as experienced as you when it comes to materials writing and I have a limited access to the tools needed to produce good content because I cannot afford to subscribe to paid corpora etc. At the end of the course, I asked student for their feedback. It was clear that they also found the course to be poorly structured and generally had problems organising all the handouts which lacked the superficial yet important characteristics which we are all used to (chapters, sections, page numbers etc.). The whole issue of copyrights perhaps deserves a separate discussion. After that course, I decided to restructure my approach, I found a coursebook which focused an the skills and systems that my students wanted to practise during our lessons (this was the original starting point for the course I just described which was based on feedback from students who wanted to get more listening and speaking practice). It was divided into neat chapters and followed a regular pattern of tasks. This coursebook comes a s a pdf and it was my decision which units I would use (based on my students input regarding topics). I supplemented it with a grammar book that organised its ‘nuggets’ around frequency rather than grammar structures. Students received folders which they used to organise and keep all the handouts ( I provided a hole punch after each lesson). In a way, you could say that the students ‘created’ their own coursebooks in this way with different sections and banks in it. This time the feedback was positive and students seemed to have a better sense of achievement.
    To conclude, I just wanted to say that perhaps students expect a coursebook and – if they do – the role of the teacher might be to meet these expectations. The points discussed in this post are very interesting and informing for me as a teacher but – in my opinion – may not be shared by those on the receiving end who may have somewhat different presumptions of what they want from the course they decide join (even though these ideas may not be based on research or even go against it). If I am to help them learn, I believe I also need to pay heed to those presumtions (even if I may not completely agree with them).

  • MuraNava

    hi Andrew(?)

    “i think the swimming analogy is rather you learn to swim by swimming; it is not the case that any instructions a swimming coach gives will affect your swimming as such – it may speed up your learning : )

    you say “The lesson for teachers and learners is that you have to make use of the language you have ‘learned’ and do so repeatedly over time.”

    one of the arguments is you acquire language by using it (you learn swimming by swimming), and that TBLT/Dogme is closer to this meaning of use than coursebooks

    from what i gather your approach is usage-based? i.e. learn from exemplars of language, a process Ellis describes as “from formula, through low-scope pattern, to construction”?

    • Lexicallab

      I’m sure Andrew will pop up at some point and comment further himself, Mura, but the swimming analogy that Geoff Jordan used was something we’d talked about a fair bit, so I’ll chip in with my tuppence worth. I think the point is that people don’t JUST learn to swim by swimming. If they did, parents wouldn’t have to spend a small fortune and invest so many hours in getting their kids through the trauma of swimming lessons. As someone who has two small kids who are both still learning, I see the direct impact that the classes – and the teaching – have on their confidence, ability to do specific things like put their faces underwater, move out of their comfort zone, etc. Anyone who claims the lessons don’t affect the actual swimming simply hasn’t watched kids go through the process. My dad exists at the other end of the spectrum: war child, never learned to swim, terrified of water . . . and never benefited from lessons and exposure to controlled entry points into the craft!

      In terms of the argument about acquiring language through using . . . well, as Andrew says, you can meet new language through using, and an attempt can be made to teach it to you at the point of need, but obviously for real acquisition to occur, this one-off encounter isn’t generally going to be sufficient, so there then needs to be repeat encounters, recycling, etc. And that’s true whether you meet an item in a coursebook or through direct communication.

      It’s clearly not the only way you can acquire language, as anyone who’s ever learned a language will know! Language can also be at least formally learned via classrooms, coursebooks, self-study, etc. and then attempts can be made to use new items during communication. There then needs to be repeat exposure, etc. as stated above.

      To pretend that somehow only one (or indeed the other) of these ways is correct just seems to be wilfully perverse.

      Not au fait enough with what exactly that Nick Ellis quote implies to really comment on the degree to which what we believe matches it, but we’ve been taught and learned enough to be skeptical of dogmas.

      • MuraNava

        Hi Hugh
        I think when most people think of having “learned” a language they mean they can use it well spontaneously, which is taken to mean the process involved is implicit. So although we can learn some aspects of language non-implicitly (declaratively) from teacher explanations, coursebook explanations and the like, most would agree this is not the goal?

        Using the swimming example also brings in issue of skill – that is doing something quickly and accurately. Cognitive science research on skills show that skills are not very transferable unless the skill domains are very similar. Hence learning to swim is different from learning to climb. In both you get expertise by doing the relevant activity not being told about the relevant activity. Your example with your kids may point to them having the opportunity to try out any swimming coaching as they are in the water. So similarly a classroom or book could provide similar opportunities when students are using the language.
        Coursebooks if you accept the above are further away from the goal of using the language than TBLT say.

        Going to what I think is Andrew’s main point – whatever applies to CB applies to other teaching issues – is not strong if we accept that CB reduces the chances of using language (i.e. primarily for meaning).

        I mentioned Nick Ellis since there was reference to recycling, reuse etc which is familiar from “frequency” approaches such as in some usage based accounts.

        • AndrewWalkley

          Thanks for the comment. To a large degree I think we can agree, which is actually my main point. In the end, because we are all passionate about what we do and see value in the way we do it, we can get into an argument that gives the impression that other ways of doing things have no value. This applies as much to ourselves as to advocates of TBL or Dogme or whatever. If you don’t mind I am not going to get in to a long comment here because I plan to do a couple of more posts which may address some of these issues. At this point I would say that potentially TBL may have more opportunities for using language, than a coursebook, but that depends on the nature of the coursebook and the nature of the focus on forms in the TBL classroom. If the worksheet that is pulled out to address an ‘emerging’ issue is a page from a grammar book, then there may be less use of the language than you might think. There is also the issue of how well tasks can be constructed to both push students language and recycle it over time. I don’t say it is not possible or a coursebook necessarily does it better, but it does rely on skill of the teacher and their ability to transfer the declarative statements of TBL theorists into effective classroom procedure. That is also the swimming. Of course it includes lost of doing but top swimmers also do lots of weight training, follow specific diets, watch films and analyse their strokes and carp swimmers like me have done a fair bit of swimming but in having little support and instruction can drown!

    • AndrewWalkley

      Hi Mura – yes it’s Andrew. I agree the instruction form the swimming coach may not lead to effective swimming and almost certainly not without doing, but speaking for our own coursebook there are certainly plenty of opportunities to do (see next post). But no instruction and only doing is not always effective either. How does the language learner start at complete beginner? Does instrauction not play a part, whether it is from a dictionary, phrasebook or teacher.
      Nor do I disagree that language can be acquired by using and that doing tasks can facilitate this, but by admission this is a slower process than if it includes instruction (if we follow the belief that it speeds up your learning rather than changing the route).
      The question then becomes, what is that instruction, when does it happen and how does the language get repeated over in different contexts (usually also resulting in greater complexity/some ‘incorrect’ use)? These are choices all writers or teachers will have to make – whether producing worksheets or lessons or courses.
      When it comes to instruction, then yes, we would suggest something which is usage-based or exemplifies what students may want to do with their language is better. Your comment and description of our ‘usage-based’ approach seems to suggest (perhaps unintentionally) a rather narrower view that precludes any freer tasks or acquisition, which is certainly not our idea.