thomas leverett weblog

Learning theory

Oddly, I was at a pleasant dinner, outside on a nice, steamy June day, watching children tear around a shady back lawn, when it hit me about how I would pull all this learning theory stuff together. Of course, a lot of it will have to be written as I go along, and is not set in stone yet, even in my mind. But I've already done a bit of writing on the subject, and I believe I'm ready to organize and move forward with what I've got. I'll lay out the plan below. That way I can say I've published it; I put it down, printed it, got it out there in the world, whether I die tomorrow or not.

Krashen was right about a number of things, and wrong about a few crucial things. When he clearly defined the difference between learning and acquisition, most veteran teachers said yes! - because anyone who has taught a language has experienced the students who can learn tricks for a test, yet be unable to use them in real situations, sometimes less than ten minutes later. We teachers knew that things, lots of things, were learned but never acquired; that Krashen was onto a crucial distinction. With time, researchers sneered at him, because he was unable to show exactly how one would prove that something was learned but not acquired. Krashen in fact has a very bad name among SLA researchers for basically having unprovable hypotheses.

On this problem I'll say, I agree with the teachers; he was onto something. Just because he was unable to show exactly how it could be done, doesn't mean it can't be done. His model, in this instance at least, was quite right.

His biggest weakness was that he was caught up in the Chomsky era and assumed or believed that so much of what happened in language learning was inherent or derived from completely natural, inborn systems. In this regard he was wrong about a number of other hypotheses; first of which was a LAD, or language acquisition device, which made me laugh when I first heard about it, and still reminds me somewhat of a tumor, or maybe a heart implant. But that's not really what i'm posting about.

Because he thought of acquisition as a natural, inborn process, he didn't take much time distinguishing the differences between first and second language acquisition, which are in fact substantial. The whole Chomskian era, with its emphasis on the inborn nature of acquisition, has this weakness. The two are quite different, and all the differences are behavioral, since the second language learner has a language system already, when he starts, and thus builds new understanding of a new language on top of assumptions which are, in many cases, false.

Thus the first language learner is like a filing clerk who looks at a huge pile of files and says, I'll have to look at these for a while before I really know how to organize them- but I'll start organizing them anyway, as I look at them, because I have to have a way to process this information as I get it, and I can always change my system a little as I go along. The second language learner, however, says, it's a pile of files, it's probably organized like the last pile. I'll use that system, the one I used for the first pile, until I figure out that this one is different, or, at least until I have a better way of organizing it, and until I really believe my life will be simpler if I reorganize it.

Notice that the two filing clerks have profoundly different approaches. Notice also that, by definition, the L2 learner whose first language is Korean has a profoundly different approach to English than the L2 learner whose first language is Arabic, for example. An entirely different system of understanding the language, an entirely different set of systemic hurdles to understanding it completely and becoming fluent.

One of Krashen's hypotheses was that there is a natural order of acquisition. People will learn one grammatical structure before they'll learn another. When they tried to prove this, things got complicated; they ended up with examples that made me angry. I'd say, sure people learn A before they learn B, since you have to learn A before you can possibly learn B...but, I figured, that doesn't prove anything, and certainly doesn't prove that language learning is a natural, inborn process, genetic and as preprogrammed as getting an Adam's apple.

I think he was wrong about that. People whose first language has relative clauses will learn relative clauses at a different time, developmentally, than people whose first language doesn't have them, or whose first language has them, but avoids them at all costs. We choose to acquire what we feel we need the most- and our perception is guided by our experience. Different experience -> different perception -> different needs analysis -> different acquisition order.

Acquisition is basically when the filing clerk says, my life will be simpler if I organize it this new way. I will reorganize my data because I recognize now that a different or better system of organization will be more efficient- will better reflect the data I have- and will thus make it easier for me to communicate, and use the system.

Here are my (rudimentary, I admit) principles of second language acquistion:

Progress in second language acquisition is marked by a voluntary change in the system of the learner- the brain reorganizing itself for the purposes of creating better language and/or being understood more clearly . For example, my Spanish teacher can teach me the difference between ser and estar (ser- the existential be; estar, the locational be) and I can get it right on the quiz and the midterm, but still use ser for all be when I speak Spanish. Then one day I figure it out, and basically put a programming order in my system, one that says, basically, when you're talking about location, use estar; use ser for the other stuff. When this order is successfully put into my production system (I actually remember when I did this- though I couldn't tell you when or where), the learner can be said to have acquired the difference. We can notice that the learner begins getting it right- getting the difference and using the correct verb in the correct slot. And, it's a different time than the time when somebody actually taught him/her the difference.

The mind is aware of the need to reorganize, long before it actually takes the time and effort to do it. This is self-explanatory. We are conservative, and don't take matters of internal organization lightly.

Nobody can tell us how to organize our own filing system. They can, however, point out truths about the way the data is organized, which we, as language learners, will recognize when we see, and respond to accordingly. Some semblance of control is necessary, however, if only for the sake of pride. It's my mind. I'll change it only when I'm ready, you lay off telling me how to think.

The kid who becomes bilingual at a young age is at a distinct advantage, because he/she sets up two filing systems right away, rather than running the second one through the first. More about this later.

The baby hears mountains of data before actually having to produce anything in a first language. The adult second language learner, however, hears much less, and, as an adult, will have to produce much sooner. And furthermore, he/she hears a lot less of some things, for example commands, that kids hear a lot of. So it is possible that he/she has a different kind of interpretation just on the basis of what he/she has already heard, let alone a different interpretation based on false assumptions carried over from the first language. One that I recall, though I don't remember it being proven in any way, was a theory that the -ING form of a verb was the basic form (the most common- the starting point). If you didn't go through several years of come here, give me a kiss, go to bed, eat your spinach..., that might be possible, I suppose. I'd like to ask some language learners.

More about this later. This whole treatise will be moved, I assure you, to a permanent home on the web somewhere.

Labels: chomsky, endangered languages, krashen, learning theory

posted by tom @ 8:16 PM