01 October 2010

An important argument against the "target language only" environment.

I have never been a fan of "target language only" language environment.  I can give 101 sophisticated arguments about why I hate it, but instead I'm going to stick to a very simple one.

The other night, I was sitting in a pub, talking to a Spanish guy.  I was advising him to watch TV serieses (I'll explain this in another post soon) and I was comparing serieses and films to novels and short stories.  "Tecks," says he.  "What?" says I.  The word he was after was "texts".  He genuinely believed that a short story was "a text" and that my "short story" was an explanation, not the word.

He's not the only learner of English who uses the word "text" too much.

Why does this happen?  It's what I call Teacherese.

When we go into the target-language-only classroom, we're immediately restricted in our choice of words by what the students already know.  If we need them to know a new word, we have to teach it to them.  If we want to get onto the "good stuff" in the language, we need to avoid being distracted by classroom talk.

So what do we do about all the "articles", "stories", "poems", "letters" and "extracts" and "excerpts from" various things?  We use one word to describe them: "texts".  By using the same word throughout, we make life easier for the students.

Unfortunately, this is a word that is only really used this way in technical circles -- most particular in linguistic study.  I have had the same argument with several teachers on this, and the usual justification is that it is real English (true) and that it is appropriate to the context (in this case a language course -- also true).

However, if we accept that we learn by example and by frequency of exposure, then it follows that we are distorting the meaning by making it too familiar to the learner.  As for the context, this is something of a red herring, as we are now assuming that an ab initio language learner will have enough awareness of register variation to determine that the language in question is register specific.  But this isn't the end of it, because register analysis and awareness can only arise out of exposure to multiple registers.  For many learners, the classroom is not merely a context, it is the entire world in which they interact with the target language.

The words and phrases we use as teachers have an undeniable and unignorable effect on our students' internal model of the language.  When we talk about "texts", we skew their model away from the native one.  When we say "say again", something never heard outside of language classes, we radically undermine their understanding of grammar.  When we use Latinate verbs instead phrasal verbs or when we use the future simple instead of "going to", we are not making life easier for our students, we are teaching them incorrect English.

As Einstein said, "Everything should be made as simple as possible, but not one bit simpler."  If as simple as possible in the target language isn't simple enough, don't use the target language.

1 comment:

Anonymous said...

Really, really interesting! A related observation of mine (I teach ESL to multiple nationalities at an intensive English school) is that many "beginning" books (all in English, of necessity) would have us use English to teach nuanced expressions and grammar points of English, to students who still don't know English, thus taking WAY too much time away from helping them understand, practice, and use structures and expressions that are learnable at their current level. A case in point is the modal auxiliary verb "had better + verb." Yes, it's pretty high frequency, but it's VERY counter-intuitive (had = past; had better = future), and our textbook introduces it at a level at which the students don't yet have enough English to be taught it through the medium of English and actually understand it.