Monday, July 11, 2016

Three views of what "grammar" is and what it is to "learn grammar"

What I have to say here is related mainly to the Cognitive Dimension of growing participation. Much of what we have taught about the Cognitive Dimension was inspired by the Psychology of Language as a branch of cognitive psychology dealing especially with comprehension and production processes. What I'm emphasizing in this post--viewpoint 3 below, was always there in the background, as I was strongly influenced by Joan Bybee (_Language, Usage and Cognition_, Academic Press, 2010) in my general thinking about grammar. (I was her research assistant for two years.) Viewpoint 1 is closest to the "cultural assumptions" of Euro-Americans, both specialists and ordinary educated people. Viewpoint 2 is inspired by "formal linguistics" which I loved as a young person. I can't say it has had much influence on the GPA, however.

Viewpoint 1) What you learn is objective facts about the language (pedagogical grammar), and then you apply that knowledge in practice until your use of that knowledge in speaking becomes fluentLanguage learning is similar to other kinds of learning.

Viewpoint 2) What you learn is a mental grammar, similar to the mental grammar in the heads  of natives. You keep modifying that grammar as you encounter evidence that your current version (current interlanguage) is wrong. These modifications to your mental grammar are triggered by relatively small amounts of experience, and so they can happen in a short time frame. Language learning is different from other kinds of learning.

Viewpoint 3What you learn is simply an accumulation of unconscious memories that all stay there in your head where they first landed-- a huge number of instances of hearing and understanding something (words, groups of words, more abstract patterns of words) or instances of successfully expressing something in speech using words and patterns. Language learning is similar to other kinds of learning (but neither is like the everyday concept of “learning”).


Viewpoint 1) is strongly “speech-led”. It is about using knowledge of facts to assemble sentences that follow the rules in order to speak those sentences. I think it represents “traditional” concepts (such as the popular PPP, which stands for present, practice, produce). It is not clear where the “correct grammar” with its collection of facts, is to be found in the universe. But this viewpoint also has its more sophisticated advocates, such as Robert Dekeyser. However he himself makes the point that this type of learning has its limits and needs to be accompanied by others. (If you are interested, see Robert Dekeyser, 2015, “Skill Acquisition Theory” in Bill VanPatten and Jessica Williams, eds., Theories in Second Language Acquisition, 2nd ed., Rutledge). Learning, if it happens, shouldn’t take that long, for any given fact of grammar, nor should practice applying that knowledge take very long.

Viewpoint 2) has been the linguists’ favourite. In this view, “evidence” (bits of language you are exposed to) acts as a trigger, and rapid changes take place in the internal grammar. Like viewpoint 1) what is learned is a recipe for assembling sentences. One problem is that theories of grammar have changed a lot, and this viewpoint still works with some theories, but perhaps not with the most influential theories. Still, a very recent article argues for this idea of rapid triggering in “grammar learning” (if anyone is interested--VanPatten, Bill and Smith Megan, 2015. “Aptitude and grammatical sensitivity and the initial stages of learning Japanese as an L2." SSLA, 37, 135–165; but also Bley-Vroman, Robert, 2009,”The evolving context of the fundamental difference hypothesis,” SSLA, 31, 175–198.) You can tell by the word “trigger” that what is learned is assumed to be learned with relatively small amounts of experience (brief amounts of time).

Viewpoint 3) is called “usage based” or “instance based” or (with a difference in emphasis) “construction based” (and other things). The idea is basically that every time you experience a word or a “construction” (which I prefer to call a pattern) and succeed in experiencing it meaningfully, it is stored in your brain. So a construction isn’t something that is learned once through a presentation of it and then reinforced by deliberate repetition until it becomes “known” in long term memory. Rather, the first time a word or pattern is  experienced, it  is “registered” and then every additional time it occurs, the new instance is stored in your head along with all the others.  (If interested, see Ellis,Nick, 2006 "Language Acquisition as Rational Contingency Learning” Applied Linguistics,  27, 1–24.). Viewpoint 3 has important implications for learning “irregularities” (such as how we learn to say “sang” and not “singed”, etc. Maybe we can bring that up again in the future.)

Which viewpoint is most in harmony with the GPA? It is viewpoint 3. It says that “fluency” in speech and in host life in general, requires that we experience speech and other aspects of life a vast amount over several years. Besides all the evidence that this is true—that becoming even moderately host-like is a multi-year prospect-- I think this is the safest bet. The first two viewpoints count on a relatively small amount of learning doing the job.  If you really follow the GPA, you won’t risk trying to learn the language based on a small amount of experience! You may recall the GPA emphasis on frequency (in listening, talking, and also in literacy, and life in general) and on familiarity (not to mention the Iceberg Principle). These all relate to viewpoint 3.

If the “facts” of grammar are important to emphasize (viewpoint 1), then the ways we deal with  grammar consciousness raising are no worse than other ways, and arguably better in some ways (thinking of actual research on this topic). For us this mainly involves our activities of structured input, input flooding, output flooding and recored yourself for feedback. In fact Nick Ellis says that the reason some aspects of grammar are never learned by adult language learners (for example, many foreigners never get over saying things like, “My brother live in London” instead of “lives in London”) is that it is impossible for our brains to experience those features (for particular reasons), and thus if one is to start storing large numbers of examples of such constructions in one’s brain one must become consciously aware of them. And that’s what we try to do with such challenging grammatical features: strongly bring them to consciousness. Who knows, but what maybe Nick is right. But the point is, with viewpoints 1 and 2 you stake a lot on the idea that the amount of experience involved in “grammar learning” is relatively small. With viewpoint 3 you assume it is very big. If you were wrong, well, then praise the Lord for your “unnecessary" thousands of hours of connecting with people in relationships since that is what it was supposed to be about anyway! But if you put your trust in viewpoints 1 or 2, doing the implied small amounts of learning, and then it turns out that what is really needed is a massive amount of experience, well, too bad.

There is much more could be said here, but that will be it for now. The GPA favours viewpoint three: Fluency is “usage-based,” based on a “summing” of enormous numbers of similar experiences in memory. It may be necessary to do something special to raise consciousness of features of grammar that just don’t seem to register regardless of the amount of experience. The verdict is still out on that, and so we do those special activities that I mentioned. We also realize that we’ll always sound foreign, and embrace that as part of “the gift of the stranger”. We aim for a high level of understanding and of intelligibility and comprehensibility, rather than native-like-ness

No comments:

Post a Comment