How do you focus on form?

<b>Forum for the discussion of Applied Linguistics </b>

Moderators: Dimitris, maneki neko2, Lorikeet, Enrico Palazzo, superpeach, cecil2, Mr. Kalgukshi2

Posts: 3031
Joined: Tue Oct 26, 2004 6:57 pm
Location: UK > China > Japan > UK again

Post by fluffyhamster » Sat Dec 19, 2009 6:25 pm

Heath wrote:Amazing how the simplest solution is often staring you right in the face. Fluffy, I think your suggestion is exactly the solution I was looking for, and one that I should've thought of straight away. I am absolutely going to try that out on the next course (we have 2 reference books that use the same approach, so we can tell them it's gotta be one of those 2). Perfect, thanks.

And I know I'm skirting the real issues here. It's a current necessity only (and my focus is purely on what's most beneficial for the trainee teachers, which, as terrible as it sounds, in some ways means I need to disregard what's most beneficial for the learners).

Regarding those issues, being a mildly stubborn supporter of a Lexical Approach (it does seem to be the most intuitive and it incorporates pretty much everything we've learned about language teaching from the past till now), when it really does come to the learners, I believe Woody's "4 is better than 2" is right - with or without the time contraints. Research tends to argue from two different angles - people learn best naturally and without technical jargon; but they learn best with a focus on form. The thing is, the 'focus on form' research doesn't really say is what a focus on form is, exactly. Number 4 (perhaps with different coloured underlines) seems to hit both, a naturalistic focus on form...
Heh, you're welcome, Heath! :) :wink:

I would assume that a (non- or not-too-technical) 'focus on form' would essentially (in the context of this thread at least) mean asking students to look at more than one instance of a particular form and/or construction (e.g. at several uses of a particular word, or several sentences in e.g. Present perfect progressive). So it would basically be a numbers/enriching-exposure game. And one relatively recent, Lexical Approach-friendly way of doing all that is of course with KWICs/concordances drawn from language corpora. (Heh, just mentioning something that I'm sure you're fully aware of but might have also potentially "forgotten" about and thus overlooked until it's explicitly brought up and into this very discussion...I know that it wasn't the first thing in my own mind right at or from the beginning of the thread, anyway, if that would make you feel any better!).

The only problem with enriching the input in any quantitative sense however is that, as we probably all know, it can lead to students overusing the form, getting hung up on it etc. So I think one has to make absolutely sure that the qualitative aspect is absolutely nailed right and at the same time, so that whatever ratio of time and examples spent in class on the form isn't at all wasted or counterproductive (that is, it will be surely be harder for any student to overuse or rather misuse a form if the actual taught example/~s that they could end up spouting is/are "at least" functionally very clear and appropriate!). One way to "get it right" is to search corpora like the BNC in the form it's been made available at BYU ( ); FWIW, below are a few tips I once wrote about it for someone:
It's really useful when you're wondering which verbs are actually used in, and thus might best exemplify, a particular tense-aspect construction; that is, it allows one to search the corpus by means of Part-of-speech (POS) tagset tags, which you can enter in a string. Beats racking one's brains/intuition for compelling stuff!

At first it's a bit tricky to operate/enter the POS tag selections, so for now as an example you might like to just copy and paste the following into the 'SEARCH STRING' box and then click the 'SEARCH' button:

will [vb*] [v?g*]

You should then be presented with a list that begins like this:

WILL BE LOOKING 223 (number of examples)
etc (in descending order of phrase frequency)

(This is an example that I gave somebody who was wondering how to contextualize "Future Progressive" ("obviously" LOL)).

You can click on the phrase or number to get a listing of the actual examples in which they appear.

If you like, I can help you work out how to operate at least this function of that site, but you can probably do so for yourself by e.g. clicking on the ? help mark to the right of the POS LIST, which provides the following tip: 'Probably the easiest way to use part of speech tags is by selecting them from the drop-down list (click on [POS LIST] to show it). By default, the tag will be inserted at the end of the string in the WORD(S) field.' (NB: One selects more than one POS by left-clicking on the light blue area outside the POS LIST box after the first selection has been made and then clicking back in the POS LIST for the second and so on).

Posts: 108
Joined: Tue Aug 18, 2009 3:38 am


Post by Heath » Mon Dec 21, 2009 1:56 am

Interesting that you mention concordancing. I have experimented once with a (single) training session on that, using both BYU and a similar one called Lextutor, and some ideas from (I think) Materials Development in Language Teaching (Tomlinson).

I love the use of concordancers in language teaching myself, but felt that only a couple of trainee teachers were able to get much out of it in that short course while the rest were looking blank and confused. Would be a great idea if they had more time to absorb it all, experiment a bit, and to have a couple more sessions on it.

The search string tips will really help me personally, though... I always fumbled around with it a bit when I was looking for more than word + ? on either side (for example, when trying to find 'will be going to' the best I could do was search for 'will' with 'going' within 2 words to the right).

Posts: 1303
Joined: Sat Jun 19, 2004 6:14 am
Location: London

Post by woodcutter » Mon Dec 21, 2009 2:26 am

I just mean that interpretation of research will tend that way in general - obviously there are those who argue for metalanguage and deep grammatical understanding. Unfortunately ESL lacks killer experiments which clear these things up - you are left the with subtle interpretation of many diverse things by people who are swayed by intellectual fashion.

Is your basic problem that you want to ground what you teach, which sounds fine, in some impressive sounding theory? A trainer must be under pressure to do that, but I'm afraid it is all a bit of a game. I do not believe that the main series of books have responded to theory that much, and have kept in the metalanguage down the years through inertia and commonsense.
(1 and 2 at the back , 5 at the front!)

Posts: 3031
Joined: Tue Oct 26, 2004 6:57 pm
Location: UK > China > Japan > UK again

Post by fluffyhamster » Fri Apr 09, 2010 1:41 pm

Thought I'd bung this in here rather than starting a whole new thread!

In the following, I'm basically on about the meta-labelling (or not) of BE in verb phrases (which can contain up to two instances of BE): ... 915#847915

Post Reply