Tag Archives: Guy Cook

In praise of non-native speaker teachers part four: Translation

Used wisely, translation can be one of the best weapons in the non-native speaker teacher’s armoury. Yet whilst it may have been undergoing something of a renaissance over the last few years, translation has certainly not always a good rep in ELT. Indeed, my own path to recognizing its potential has been a long and winding one. Back in 1993, when I did my four-week CELTA course, there was certainly no mention of it, and in the two main bibles that I read at the time in order to glean ways forward – Jeremy Harmer’s PRACTICE OF ENGLISH LANGUAGE TEACHING and Jim Scrivener’s LEARNING TEACHING – there wasn’t much to get me thinking about translation either. In the latter, there was no mention of the phenomenon at all, whilst in the Harmer, I was told it was “a quick and easy way to present the meaning of words,” but then immediately warned that it was “not without problems” – it’s not always easy to translate words, and even where translation IS possible, it may make life a bit too easy for the students by discouraging them from interacting with the words.

Having not learned how to make life easy for my students, I set off to a monolingual school in Indonesia to get started on my teaching career – and quite soon I started noticing a strange thing happening. Students would ask me what a word meant, I’d go through contortions to act it, draw it and explain it and after a few minutes of killing myself, students would suddenly look pleased. I’d think “Finally. They understand what a frog is and say to each other, for example, “Oh! Kodok!”

As I was learning Indonesian myself, I learnt a lot of it from hanging out with English and American friends who had lived there longer and who spoke the language better. I’d often find myself asking them So how do you say . . . in Indonesian? and essentially teaching myself chunk by translated chunk. I also started slowly realising that a lot of the problems I was having were down to having learned a word and thinking it’d always work the same in Indonesian. I learned, for instance, how to say in Indonesian to my low-level classes OK. Let’s check the answersMari kita periksa jawabannya – and so logically assumed that the Indonesian word periksa must therefore be equivalent to the English word check. However, when Indonesian friends came round for dinner and I told them Saya akan periksa makanannyaI’ll check the food – they’d laugh and correct me and say coba makanannya – which for me meant try rather than check.

Once back in the UK, I noticed the same thing the same thing happening in reverse. In classrooms, I’d frequently be trying to elicit a missing word – say, for instance, here:

He’s got a really good job. He ………… a hundred thousand a year.

and students would shout out WINS! WINS!! and I’d end up retorting “Maybe in Portuguese, yes, but in English anyone?”

As time went by, I also started to recognise very common mistakes from certain language groups of learners, which I realised must be down to poor word-for-word translation, so my Japanese students would say I was stolen my mobile / bag, while Spanish speakers during tutorials would enthusiastically report that was a course very interesting.

So as you can see, I’d spent many years skirting round the fringes of the translation in language teaching issue, but had never really paid that much mind to it, if truth be told. What really made a convert of me was actually one little feature we wanted to include when writing OUTCOMES – a section called Language Patterns. The idea was that we somehow wanted to focus on lexico-grammatical patterns that weren’t strictly grammar, but that were definitely beyond single words – the kind of thing you can see below:

Mongolia is known as ‘the land of the horse’.

Shanghai is known as ‘the Paris of the East’.

Aubergines are also known as eggplants.

The area is known for its oysters.

The village is well known for its leather goods.

This rare species of shark is known to inhabit fresh water.

Very few details are known about this rare species.

And we wanted to encourage teachers to get students to notice them. Now, you’re all undoubtedly aware of the importance of noticing – it’s been central to theories of how language is acquired for over twenty years now. Back in 1990, Schmidt stated that while noticing does not automatically guarantee acquisition, it nevertheless remains true that features of the language cannot be learned UNLESS they are first noticed. Schmidt was talking more about structural grammar in its traditional sense, but Rod Ellis went further in 1997 and stressed the importance of drawing students’ attention to items that do not conform to expectations and that may therefore not otherwise be noticed.

Noticing is so central to learning that you could quite easily claim it is one of only four or maybe five things that needs to happen for any item or structure to be acquired. Essentially, to learn a language people need to:

• hear or see the language

• understand the meaning of what they hear or see

• pay attention to the language and notice aspects of it

• do something with that language – use it in some way

• repeat these steps for the same language repeatedly over time

The question was, though, what was the most useful way of trying to encourage noticing when space in the book was limited and when these were not the kind of core structures that teachers expected to find in the book. Was it enough to simply sort structures, show them to students and ask them to ‘notice’ the patterns? What might encouraging noticing actually involve and how could a teacher say with any degree of certainty that their students had noticed?

As we were to find out, facilitating noticing in class proved far more problematic that we’d initially anticipated. Initially, our rubric for these sections was simply Which patterns can you see in these sentences? Now, you think about how you might answer that question with particular reference to this particular set of language patterns from OUTCOMES below:

It’s hardly the same thing!

Hardly an instant solution then!

It’s hardly surprising people are concerned about it.

Hardly a day goes by without hearing one of these stories.

I hardly know anyone who agrees with it.

There’s hardly any funding available for research into it.

What WE noticed when we asked students to do this and to then share their insights in pairs or groups was that (a) they didn’t actually notice all that much and (b) it was hard to verbalise whatever awareness of underlying patterns they might’ve become aware of in this manner. Even if both of these barriers were overcome, there was then still the nagging doubt that none of this would lead to better production; that the noticing would all essentially be in vain. We then tried translation and in one particular class I had my eureka moment. Now, again, you might like to stop here and try the previous exercise, but this time with the following rubric: Write the sentences in your language. Translate them back into English. Compare your English to the original.

I had a French student in one class, who spoke very well, but often in a kind of French-in-English way, and who was also very resistant to the idea of using translation. “But I understand it all,” he would protest. “There’s no need!” “Please!” I would beg him. “Just do it for me!” “But it’s the same in French,” he would try to persuade me. “It can’t be,” I’d point out – “for starters, it’s in French! Please! Just to shut me up, try it.” So translate he did. I then kept the translations and the next class I pointed at one of his translations almost at random and asked if he could say it in English. “Of course,” he replied. “It’s Hardly a day is passing without that I hear about one of these stories.”

“Ah-ha!” I suddenly screamed. “That’s the FRENCH pattern, but you haven’t noticed the ENGLISH one!”

Translating back and forth between languages like this forces noticing in a way that nothing else does. So why, I started thinking, don’t more teachers do it? The bulk of classes around the world are monolingual with relatively bilingual teachers. And many of us who are proficient to at least some degree in two languages code switch all the time – with friends, relatives, lovers. It’s the norm rather than the exception.

Yet monolingual teaching has come to be seen as the norm, as the most desirable model! However, as Guy Cook points out in his quietly furious tome Translation In Language Teaching, the reasons behind this dominance owe far more to commercial and political imperatives than to science or pedagogy! How can this surreal state of affairs have come to pass? And how have so many teachers who could potentially benefit from a world in which their language skills were allowed fuller expression been brainwashed into believing they have to try and emulate the sad, sorry islands of monolingualism natives so often find themselves on?

In many ways, I fear, we are STILL suffering from an ongoing backlash against grammar translation, a backlash that has gone on so long and been reiterated so mindlessly that it’s become almost a subconscious knee-jerk state of mind. Grammar Translation was very much the dominant mode of language teaching right up until the tail end of the 19th century. Rooted in the teaching of Greek and Latin, with which modern languages vied for respectability, the emphasis was very firmly on writing, on grammar, on accuracy and on the ultimate aim of allowing the student to read literary classics in the language they were learning. Grammar Translation is what people often imagine either when thinking of traditional approaches to language teaching or else simply to translation in language teaching in general. As well as learners memorizing huge lists of rules and vocabulary, this method involved them translating whole literary or historic texts word for word. Unsurprisingly, new methodologies tried to improve on this. The Direct (or Natural) Method established in Germany and France around 1900 was a response to the obvious problems associated with the Grammar Translation method. In the Direct Method the teacher and learners avoided using the learners’ native language and just used the target language. Like the Direct Method, the later Audio-Lingual Method tried to teach the language directly, without using the L1 to explain new items.

The Reform Movement, which was the initial reaction against Grammar Translation, placed the primary emphasis on speech, and generally insisted on an English only approach, but still allowed some translation. These ideas were picked up and simplified – and then codified – by schools during the first great language teaching boom and Berlitz, founded at the end of the 19th century, insisted on natives only, speech only and no use of L1. Indeed, translation became a sackable offence. This led directly to the pillars of practice that haunt us to this day: monolingualism; naturalism – the idea that learning L2 can somehow mirror the ‘natural’ way we learn L1; native speakerism and absolutism – the belief (or claim) that Direct Method is the one true path!

Subsequent so-called ‘humanistic’ methodologies such as the Silent Way and Total Physical Response and communicative methodologies moved ever further away from L1, and from these arose many of the contemporary objections to translation. Sure there was the odd exception, such as Community Language Learning in the 1970s, which accepted the whole human range of approaches, including negotiation between student and counselor teacher, within which translation was seen as one tool among many, but such approaches were few and far between.

All of which brings us to our current state of play, where countless – and often groundless – fears abound: students will end up using L1 all time, when aim is use of L2; the skills involved in translation are not suitable for all learners – and may only suit those who are analytical, older or better; learners may not see the value of translation value or only see it as hard or specialised; it’s hard to set up and run in class; it requires extra motivation from students; it needs a teacher with a good knowledge of students’ L1 and culture and thus doesn’t work in multilingual classes – and on and on it goes.

At its worst, anti-TILT (Translation in Language Teaching) rhetoric is rooted in dialogue focused on monolingualism and the supression of other languages – as can be seen in the States at the moment, where folk proudly sport Speak English or Die T-shirts and where a recent airport best-seller is entitled His Panic: Why Americans fear Hispanics in the US.


Yet as I hope I have already persuaded you, there are many strong reasons in favour of using TILT. Some of the strongest are actually evidence based. For instance, in a 2008 study, Laufer and Girsai taught vocabulary to three groups using three approaches – meaning-focussed, form-focussed without translation and through contrastive analysis and translation. Both passive AND active retention was way higher with the third group.

Translation is, by its very nature, highly communicative and is a real world activity for the vast majority of students at some point in their language-using lives. On a more meta level, you could almost argue that translation makes the world go round – the UN, the EU, business, academia, and so on all rely on it.  Whether we like it or not, the process of understanding L2 by looking for L1 equivalents has always been a frequently used strategy for learners. If you accept this, then there comes a need to develop it in the right way – to hone it.

Lower-level students use translation all the time – and for higher-level learners, it’s almost by definition what it means to be good! I’d be amazed if I were to go out for dinner tonight with any of you reading this and found that you were unable to translate an L1 menu, for instance!

In terms of student-centeredness, many students – especially younger ones and those at lower levels, though perhaps not only them – look more favourably upon bilingual instruction and, therefore, translation than has previously been admitted.

Irrespective of all arguments in favour of using TILT, the bottom line is that it’s the most effective way of doing stuff that needs to be done! In many ways, as well, translation is one of the most authentic tasks that we can engage in in the classroom as it’s something we all do all the time – in the so-called real world. There’s also the very real possibility that for many students, translation will be the main – or maybe even the sole – activity connected to English that they engage in later in their lives!

In addition to everything else, it’s a time-efficient way of dealing with such time-honoured problems as false friends, it requires minimal preparation – and, let’s be honest, the recommendation that foreign-language classes be taught exclusively in the foreign language remains, shall we say, ‘aspirational’ at best!

To those of you who STILL remain sceptical, look at it this way. From L2 to L1 is less an absolute act and more just part of a spectrum. When we explain new language in simplified language or with gestures, we’re already engaging in a form of translation! Given this, surely it should not be too much of a leap to then allow the principled use of L1?!

Henry Widdowson once said that that the error of monolingual teaching is that it misunderstands how learners of English engage with their new language, and the purposes for which it is being learned. He warned that to proceed as though the learners’ own languages do not exist, attempting to induct learners into a local monolingual native-like perspective, is to profoundly misunderstand what is happening. Learners will ALWAYS relate new language to their own, even if only in their own minds, and if forbidden to do so, will nevertheless continue the practice as a means of resistance!

In short, humans teach and learn by moving from the familiar to the unfamiliar, by building new knowledge onto existing knowledge. Language learning is no – or should be no – exception!

Interestingly, the grammar-plus-words model of language that still prevails in many coursebooks works least well with TILT. What works best is collocations, chunks and patterns. Lexis, in other words. What clearly rarely works at all is single words – and, to a lesser degree, grammar, especially if we’re looking for direct equivalence, though as I said earlier, it can still be useful to understand L1 transference errors.

This does all seem to suggest, then, that if we are to get the most from TILT, then the time has come to drop the dominant model of grammar plus structures and to embrace instead an approach to language that sees grammar and vocabulary as inextricably intertwined and contextually bound.

So what kind of activities can we do that might take all of this on board? Well, to close, here are five that I have done in recent months – and that you might want to try for yourselves. I should add that I work with multilingual groups in London, and have still found these tasks work fine. I expect that many NON-natives working in monolingual contexts where students share their own L1 have plenty of other ideas on how translation might be fully exploited – and I hope to read more about these in the comments section below!

1 If you come into class and students are chatting in L1, get them to write the conversation they’re having first in L2, and then translate it. Help them with any expressions they’re not sure and maybe, if you can, round up by pooling a range of new expressions / chunks that have emerged through the process of translation on the board.

2 When students lapse in L1 during freer speaking activities, note this down and then during your round-up either give or else elicit English versions.

3 Give – or ask for – translations of single words as a STARTING point, but then show ways in which these words are NOT the same! So say, for example, the sentence I’m responsible for hiring and firing comes up, you might want to say the L1 equivalent of responsible, but then say that in English, you‘re responsible FOR doing something, not responsible of.

4 Allow students to translate things that they may have to translate in ‘the real world’. Use L1 as a resource and as a bridge to L2. As Guy Cook notes, there are countless possibilities here that lend themselves to communicative / task-based activities: a company entering negotiations with foreign partners may receive documents and communications which first need translating by bilingual staff; evidence in a court case may need translating before a judgement can me made – or, as in this exercise from OUTCOMES intermediate, a menu may need to be translated before diners can decide what they want – and don’t want – to eat.

Conversation practice

A            Write a typical menu for a restaurant in your country. Write it in your own language.

B            Work in pairs. Imagine you are in a restaurant that does not have an English menu. You are trying to decide what to eat.

Student A: you are visiting the country on holiday or on business. You do not speak the local language.

Student B: talk Student A through the menu.

Student A: reject at least two of the things on the menu. Explain why.

5 As a self-study device, make students aware of things like the Word Reference forums for bilingual learning communities and encourage them to use them.

Looking forward to hearing your thoughts, questions and comments on this paper.

Advertisements