|
We're not kids. Adults benefit from explicit grammar instruction--actually, so do kids after about 3rd grade (developmental psychologists have some explanations for that because linguists don't).
Pronunciation starts being affected after about age 6 and is difficult to make native-like after age 12-13, grammar's affected at age 12-13. Get them by those ages or you'll almost certainly be marked as non-native forever.
I think kids also memorize until they've internalized new words and constructions, it's just that they can do that fairly quickly. So kids talk to themselves, often repeating words and expressions they've heard. It's not memorization as we usually understand it, but they remember the words short-term and practice them until they have them long-term. It's a question of taking stuff we've "memorized", learned explicitly for explicit recall, and internalization, automatization, routinization, getting to the point where we don't put forth more than a trivial effort to produce the output or recognize the input. My kid was downright weird: He'd lay in bed at night and practice sounds, initial consonants to start, and then finally final ones: ka, ka, ka, ka, ka. A year later we heard him practicing vocabulary until we fell asleep: "fender, fender, fender ... crab, crab, crab ... colleague, colleague, colleague." He'd repeat the words until he got all the sounds right, and then move on.
Aural and oral skills aren't much different from reading and writing in some ways: There are layers of what's to be learned. My currently favorite theory of phonology is Joan Bybee's. We remember all the "tokens" of a word that we've heard, with some sort of context and information about them: "can", with the 'k' more or less aspirated, the 'ae' more or less nasal, shorter or longer, the 'n' more or less clearly uttered. We construction a mental "ideal" representation, some sort of weighted average, and then compare the incoming speech stream with the various tokens we've heard. We hear the low-level phonetics and glean contextual information from that token even as we record that token, so if you change dialect areas your "ideal" token can drift. This requires initially setting up a sort of "mental space" for the word, of course, and that's what happens when we learn a word. In any event, you still have to memorize then learn something like an auditory representation. Note that in childhood second language acquisition kids have a quiet period before they start to speak, as they learn words and phrases and even often subarticulate them (practice, if you will).
When it's time to speak, we come close to the center of the appropriate cluster of tokens for speed, context, style, etc. We practice them if we're adults learning a second language, seeking to automate production. It's where I flub Russian and other languages--I don't speak them much and don't think I'll ever speak them much, and when I need to speak I have to work on pronunciation. It's not automatic, I'm sort of generating the phonetic form as I go and it takes mental effort (mostly unconscious effort, but it increases the cognitive load by a fair amount). So I can rattle off long, complicated sentences in a piss-poor accent or I can refine my accent (often by mimicking some Russian!) and lose much of my stylistic flexibility. When I've been put in a position needing to speak often, my fluency improves from "Me--Igel" to "Hi, my name is Igel" to "Please be so kind as to allow me to introduce myself, if I may--I'm known in some circles as Igel" in a matter of weeks.
The four skills (reading, writing, aural, oral) work out about the same, in principle: If you have to work at spelling or speaking or hearing or reading it slows you down. You need to practice all four, "learning" (or 'acquiring', to use Stephen Krashen's preferred term; to use mine, automating perception and output). Sometimes you get a freebie: Learning Spanish or Polish you have the alphabet learned and most of the sounds approximated early on, and most of the grammar is fairly standard West European with a fair amount of overlap with English, and getting from written form to spoken form uses well-worn pathways. Learning Arabic you get no freebies in reading/writing, and far fewer freebies with grammar.
So, yeah, we need to memorize, even for speaking and hearing. But we can forget what we've memorized after we've learned it. :-)
|