Grammar-Translation as a fall back position

In today's post I want to muse a little about the practice of using grammar-translation as a 'fall-back' position for when some type of communicative teaching 'isn't working'. I mean that in two different ways, and I think teasing out those two things will be useful.

Full disclosure, this is occasioned by hearing from a student or two, and a faculty member, trying to implement communicative teaching and having some struggles; but this is not me trying to solve their problems, just thinking through this type of problem more broadly.

Here is my starting point for thinking about this: "comprehensible input works". That is to say, it's well-established in SLA that the thing that drives language acquisition is exposure to input that is meaningful and comprehensible. Not much else seems to drive acquisition of a language. Despite our strong intuitions otherwise, that you need explicit grammar or whatever, 'teaching grammar' does not appear to lead to language acquisition. So, unless the field of SLA does a total 180-degree turn, this remains axiomatic in the considering how we acquire languages.

However, lots and lots of things can 'go wrong'. It is not the case that you can just throw students extensive reading or extensive listening or just talk to them all day in Latin or Ancient Greek, and assume that they'll succeed. CI isn't magic, and it's almost certainly the case that at some point in either our teaching our or learning that we'll hit not a plateau, but a steep incline.

What happens when you come across an incline? You slow down, because the challenge is steep, or possibly insurmountable. Now, to step out of my metaphor, what happens when the texts, visual or audio, are too difficult? It means you have to troubleshoot.

This is, however, the point at which we tend to fall back to Grammar-Translation. Why? For a few reasons, I suspect.

1. As teachers, we often have a very good explicit and external grasp of the grammar, and so we can explain how the pieces of the language fit together, ad so how texts mean. We can render thinks comprehensible, and we were probably trained pretty well in doing so.
2. As learners, we want to understand a text, and we want to understand it explicitly if we can't understand it implicitly, and so we also want to 'know the grammar', so that we can figure out texts. If the explicit grammar is hidden from us, we feel like we've been cheated of the tools. It's like we're playing a game and the teacher has modding tools and won't let us use them.
3. We feel more of a sense of mastery of a language when we know and understand explicit grammar.
4. We still tend to think that grammar will lead to acquisition.

And so, whether on the teacher or the learner side, resorting to grammar-translation seems like a fall-back position for when things get tough.

The right question to ask in this situation is if X isn't working, should I persevere with X, or try Y? So it is right to ask this. If a 'communicative approach' does not seem to be delivering the results it promised, then what needs to change? I think the problem with our answer in this situation is that it looks like this: "Well, I guess we'll go back to grammar-translation then; at least we know that works."

It doesn't work. Or, this is most important it doesn't work the way we think it does and it doesn't achieve what we want it to.

Why yes to grammar, sometimes

But before I talk about what I think one ought to do, let me make a positive case for how I think you absolutely can and should use grammar + translation and why you should teach explicit grammar.

I don't have an anti-grammar stance per se. I think, especially for anyone studying historical languages, there is real value and worth in learning explicit grammar. The important thing is to realise what is does and does not do for you.

Learning grammar and practising translation does not, as a practice, increase language acquisition.

Understanding the grammar of a language is understanding how a language works, or in other words how it makes meanings.

So, let's take a slightly different tack for a moment. What happens when you come to a sentence or text in a language that you cannot read and process and understand 'naturally'? Then it's incomprehensibly input. And it stops you.

There are only so many options. One, give up and move on. Two, get some aids to help you. Three, apply grammar and translate.

Option 2 might involve having another person paraphrase it, provide scaffolding, have a conversation in which they explain certain things, use target-language or monolingual glossing, etc. etc..

Option 3 means you have the tools to say, "okay, I don't know what this means, but I do know (a) the structures of the language and how they make meaning, and (b) I can look up vocabulary I don't know, so (c) I can translate this sentence or text into my own language and that's one way of making it comprehensible.

The next mistake comes, though, in thinking that this counts as 'reading'. Or that it is a viable way to acquire language. It's not reading, it's translating in order to understand. It might enable reading, but it's not reading. Nor is it really a viable way to acquire language. Unless you use that translation to go back and actually read and process the text again. But it's very slow to do this.

Translating into your own language in order to understand gives you the illusion that you're reading and understanding a foreign language, and precisely for that reason you should be wary of using it as a primary strategy. It means the things you are reading are not comprehensible input for you. It also can fool you into thinking that you can read far harder things than you probably should, and can reinforce the idea that if you just study grammar and vocab, you will be able to speed-translate your way to proficiency.

So, then, why teach grammar and translation at all? Because it gives you the tools that precisely when you do run into a roadblock, you can solve it. You can render it comprehensible for yourself.

To return to my initial question/prompt then: grammar+translation is a fall-back when you don't have other options for rendering a text comprehensible, and it's really good at that. It's also really good if you want to talk and discuss and analyse how texts mean.

But it's not a good fall-back in general course design for acquiring a language, because it is trying to solve the wrong problem the wrong way. The problem isn't really "these texts are too hard without learning grammar", it's just "these texts are too hard for where a learner currently is, so they aren't providing comprehensible input".

The solution to that problem is, hopefully, becoming more obvious. It's more comprehensible input. Arrgh, but that was the problem we began with - CI didn't seem to be 'working'. I suspect that most of the time when it's not 'working', it's because it's not (a) enough (b) actual CI. That is, trying to read or listen to a lot of content that is too difficult for you doesn't do you much good. You need better, more appropriate, content. There is also the 'enough' problem. I think in historical languages we generally vastly underestimate the amount of easier content people need to get anywhere acquisition-wise, and we confidently overestimate, or oversell, the ability to get pretty far in the language with not-very-much time.

If what you're trying to read isn't making sense, the problem is that it isn't making sense. If you have to make sense of it, grammar + translation will do that for you, but if you don't have to, then reading something else that's appropriate, and lots of it, is what's going to serve acquisition in the long run.