Jim Tripp, in a recent comment on a question by Andrew Snider here, wrote:
…Andrew, I think narrow and deep is best with stories, instead of trying to do “too many”. When you finish the story, have the students draw it, mime it, or some other way of piling on more CI. Then go back and add details. I think the story will keep interest better if you go deeper into the feelings, thoughts, and dialogue of characters….
I think Jim advises that we go “narrow and deep is best with stories” so that our students get more reps. We try to sneak as many reps as we can in as often as we can at a slow pace and packaged so that they are interesting, meaningful, and on a good day compelling, to our students. So we use some of the ideas from the Two Week Plan like Textivate because it presents the same content in a different way.
Yesterday in a level 1 class I spent at least 30 minutes on “was returning home” to set up a Matava story about a refrigerator. My students may not have acquired some of the stuff around that structure, like “was returning home while doing Prancersize” or “was returning home via small airplane” but I would be willing to bet that “was returning home” was acquired in the real way. Maybe.
We forget that acquisition takes place over years in the 24/7 original perfect model of language acquisition. The one Blaine Ray didn’t design. Do we not exhibit a bit of hubris when we try to imitate the original design in our classrooms by compressing it into 4 hours a week when the original design calls for 168 hours per week?
Here we all are thinking that because we have TPRS we can do a story in one class period and expect huge gains when the truth is that the original design requires a lot more time and reps than we get in class. I tend to think that if a person is exposed to comprehensible input for 168 hours per week over many years, they would tend to learn the language a bit better than a person exposed to only 20 hours per week.
It is an unreasonable expectation we put on CI/TPRS, to expect it to bring acquisition of a targeted structure in just a few days. And yet we go into our classes each day and expect miraculous gains because we have found this model. In doing so, we become like those grammar teachers who, having once “covered” the material, expect it to be have been learned.
Since learning a language is so vastly different than learning say, math, which is an activity involving conscious left brain analysis when what we do requires unconscious whole brain/right brain functioning in ways we could never understand, so that when students have had just a small percentage of the necessary exposure the targeted structure actually requires, there are no, or very limited, gains.
Jim has long written here about the need for sneaking in as many reps as possible, and he often brings us back to this necessary but often forgotten need in our work, the need for far more reps than we think are necessary.
I would go so far as to say that if I were to do my job right I would probably need to do one story per month, and that would be the fast version, that story being presented in a million different ways. THAT is the model that most closely mirrors the original design.
The oft repeated lament about “they forgot what I taught them” is not that they forgot at all, but that they never acquired anything in the first place, because they only got, say, 2000 reps on a particular structure when they needed perhaps 28,000 for actual full acquisition. How we fool ourselves!