Don't correct. Re-write.
Steve Kauffman put up a blog post yesterday: The role of correction in language learning - very limited and only with permission. His basic claim is that corrections don't work very well. People don't remember the corrections you made. Even when someone does intellectually understand a structure that they're messing up, they still may not be able to produce the correct version when the pressure's on. And frequent corrections interrupt the flow of conversation and potentially de-motivate the person.
All good points, but I want to propose an amendment. Assume that someone has come to you looking for corrections. Assume that you're not interrupting them. (I agree with Steve's suggestion that written corrections are better, because they're easier...
Here's a little example of something English learners get wrong that I attribute to a failure of textbooks and teachers to properly reflect the way language is used.
I've been spending a lot of time recently hanging around Lang-8.com and helping people out by correcting their journal entries. Most of the corrections are novel - they're things that I've never encountered before. Others are so common that I barely even think of them - problems with plurals and articles being the main example of this.
There are a few types of mistakes that are common enough that I've developed a standard response to them but not so common as to seem hopeless. One of these is the over-use of "will". Here's a made-up (but not too far from what I've seen) example:
Tomorrow I will go...
This morning I was trying to think of some phrases that use the word "face" to show to someone and did a quick Google search. One of the results I came across was this: http://www.phrases.org.uk/bulletin_board/32/messages/937.html
I think that this is fairly typical of phrase lists and repositories that you'll find on the web - long, exhaustive lists that contain the best and most useful phrases right along with ones that you've never heard of ("A face like a fiddle") and those that are commonly known but little used ("Cut off your nose to spite your face").
I've written before about the need to differentiate the everyday phrases from the merely interesting ones. A longer list of phrases isn't better than a short one. It's worse, because the language learner has no way to tell...
I was going through a list of English phrases that I collected from watching TV, reading online articles and so on. I came across this one:
Are we still on for Saturday?
As natural an English expression as one could hope for, but not something I expect to hear from my students or friends who didn't grow up speaking English. And certainly not a phrase I would expect to encounter in a textbook. This is exactly the kind of language that I love to find and point out.
I want PhraseMix to point out the uncommon common phrases. "Common" because they're used all the time; "uncommon" because it's rare that anyone teaches them.
Ever buy a book of idioms or phrases in the language you wanted to learn? Somewhere at my wife's parents' house in Japan is a book that I bought several years ago with several hundred pages of phrases. I tried studying some of them, but at some point tried out a few on native Japanese speakers. The response I got was,
"No one says that."
Not being a native speaker myself, I didn't have the facilities to judge whether this was an accurate claim. It may be that the Japanese people I associate with are not particularly literate. Or maybe, in their enthusiasm to find enough phrases to make a book out of, the authors ended up including a lot that were not very common.
But - me and my language conspiracy theories - I have another explanation for why my friends may have...