Eric wrote this over on the Forum. It nicely covers the targeted vs. non-targeted discussion and introduces a term we should all know, – “fluency development strand”. So your homework is to read this article and get command of those terms if you don’t already. There is a category on “Non-targeted” if you want to go deeper with this, with articles in it since 2008 when Dr. Krashen started publishing more frequent articles about targeted vs. non-targeted vocabulary. If you already know about these terms, you don’t have any homework tonight. And don’t forget you can get graduate recertification credits easily and cheaply from Portland State University for the reading you do here. See that category as well.
Here’s Eric:
If you want to read an eclectic framework for teaching language, then check out Paul Nation’s recommendation.
http://www.victoria.ac.nz/lals…..trands.pdf
25% meaning-focused input
25% meaning-focused output
25% fluency development
25% language-focused learning
Paul Nation responded to my question – Which type of comprehensible input do you believe is optimal: targeted or non-targeted?
“It seems what TPRS is doing is what I would see as part of the fluency strand. What Krashen recommends [non-targeted CI] is part of the meaning-focused input strand. Both are needed. So for me it is not a choice between the two but how you can have both.”
So, under the Nation model, you’d do both! When we stick to just known language (98%+), we are working within the fluency development strand. That is “targeted CI.” When there is 95-98% vocabulary knowledge of input, then we’re working the meaning-focused input strand. That’s non-targeted CI.
He designates 75% of the time to meaning-based activities. To be fair, we also spend a tiny fraction of the time, consciously translating structures and any out-of-bound words. This is a guy that wants a little of everything. He supports so many theories (input, output, skill-based) and apparently doesn’t see them to be conflicting.
It’s so impossible to read SLA research, because they say something was effective and efficient, but don’t tell the reader how the study measured the gains (communicative ability? Monitor-free? Delayed testing?).
And Nation gave me this gem in response to another question – Is it best to build listening vocabulary before students read?
“simultaneous listening and reading is fractionally better for vocabulary learning than reading by itself (Brown, Waring and Donkaewbua) and these two are much better for vocabulary learning than listening by itself. Having reading wait until there is enough listening input seems unnecessarily restrictive. For learners of a language, listening speed and reading speed are probably not too different. Eventually reading speed should be substantially faster. I guess my general theme is that a balance is better than restrictive choices.”
And he’s telling me that vocabulary acquisition is much less effective from listening than from reading. Then again, when a study measures vocabulary gains as translation tests, then we don’t really know if those words can be used in real-time communication. And apparently he’s not against early reading. Nation does favor deliberate learning of the high-frequency words (2,000-3,000 word families) during that 25% of the class and 75% of the class using those words in meaning-based contexts.