Everyone knows that grammar-based language learning (of the kind traditionally taught in “foreign language” courses) doesn’t work, and that immersion methods are better. But currently there are at least two different immersion strategies:
- Comprehensible input aka Krashenian: Input before output; huge quantities of self-selected, compelling input; long silent period; everything done at the pace of the learner, and at will. Besides Krashen, this is the method of AJATT and Antimoon & their sphere of influence. It’s also the method of everyone who, like me, learned a language by reading books or playing games.
- Interaction-oriented: Michael Long’s theory; outside the ivory tower, the method of “Fluent in 3 months” guy. Put yourself in a situation where you must interact with people in the L2. No input/output separation, emphasis on spoken language conversations with feedback.
I’m thinking the relative merits of these strategies are more apparent when we think of language as a means, not as a goal (in other words, when we stop trying to learning a language and start using it to do things at once, even if poorly at first). If my uneducated guess is right, couchsurfers and immigrants who want to socialize with other people would benefit most by socializing with other people, while philologists and lit students would probably do best by staying at home with their dictionaries.
I think the master rule is the “keep in contact” metamethod that’s been proposed by a lot of people (Chomsky, Krashen, Khatzumoto et al): You have to do something that 1) you want to do and 2) needs the L2. The theories and methods are of secondary importance to this; no matter how well-researched, if you find the method boring you’ll give up midway. An important point that I missed at first is that, if I wanted to do something just because it would improve my L2, I probably didn’t want to do it strongly enough. The desire to learn a language in itself is a very fine thing & I admire it immensely (—as in, it’s probably a deal-breaker in my heuristic to evaluate sexual partners), but at this conscious level it’s really a desire to learn about the language (that is, it’s the impulse of a linguist). It isn’t enough for the same reasons that the academic desire to learn about the philosophy of martial arts isn’t, in itself, enough to become a fighter. We might be curious about a few hundred quirky words or funny kanji, but I think no one actually want to drill & memorize ten thousand combinations of the same articulated parts. The best motivation to learn the ten thousand is not wanting to “own” the language like a prize trophy, but wanting to crash through it; it’s when you want so hard to read this specific book & language is a barrier getting in your way.