Sunday, February 28, 2010

Na'vi and simpler approaches to language learning

Since Tolkien, I've been intrigued by conlangs, though not enough to actually learn one. I'm leaning the same way on Na'vi, the language of the blue people in the movie Avatar. However, I'm impressed by the way the thing is being marketed.

When you search for information on Tolkien's languages, most of it is pretty in depth. I've never had an interest in learning Klingon, but it seems to me to be something for pretty serious hobbyists. Na'vi, though complex and idiosyncratic - infixes and lenition? - is being presented as something fun to learn and chat in. The curious can visit a site,, where there are downloadable guides, and even a workbook. It is suggested that you print out two copies of the workbook, do it once to sort of learn and do it again to review and solidify your learning. And when you open the workbook, it's full of word-searches, crossword puzzles and fill-in-the-blank exercises, not baffling grammar tables.

I bring up the Na'vi matter for two reasons. First of all, if you've always wanted to learn a conlang but weren't sure where to begin, here's what looks to be an easy one to get started in. The second thing that interests me, though, is this workbook approach. Have we been going about language learning all wrong? We know that with Iverson lists, SRSs and Pimsleur's Graduated Interval Recall, learning, forgetting and relearning is key. What if the answer is not to find the perfect course, but to find a good enough course, rush through it to get the main idea and do it again? I recently got Unforgettable Language's Spanish vocab program and that's what they suggested - listen to the program in a weekend, then listen again the next weekend.

I've had out the Quick and Dirty Guide to Learning Languages again, and made myself a simplified version of the vocabulary section to put in for Uzbek. And my thought has been to fill it in once with research, review for a week, then see how much I can fill in from memory and how much I have to look back for. Rinse and repeat till I know 90%. But this has also set me to thinking about how far one might get by, for example, learning the contents of a simple phrasebook, learning enough grammar to see how the phrases work, then relearning those contents. I'm now thinking about better ways to make an Uzbek book - for myself - along the lines of the Na'vi workbook. After all, why spend your time just learning languages when you can spend it looking for easier ways to learn languages, eh? ;)

Sunday, February 21, 2010

Fumbling Toward Polyglottery

This week, I've been sort of scattered, language learning wise. Part of this ties into language learning goals. And part of it, let's be frank, ties into ego.

I've always enjoyed dabbling with a number of languages. But I've always been stuck at two - English and French - where I have solid competence. On the other hand, I've let my Spanish and Italian wax and wane. The other day, I was updating my profile at and I noticed the criteria for basic fluency. Modestly estimating my abilities, I'm just a little bit short for Spanish. And so, while this is Italian month, I've wanted to dig in with the Spanish enough to maybe count myself a weak triglot. (My Spanish is far enough ahead of my Italian that this is the way to go, and, besides, I talk to Spanish speakers in Spanish all the time so ramping it up a bit will just make it easier to do something I'm already doing.)

This week, then, I've slogged through a bit more Michel Thomas Italian. I'm partway into disc 6. And I've worked through lessons 1-4 of the DLI's 200 Hour Uzbek course (though I don't have the audio, so my work consists of working through the written exercises and making sure I know all the vocabulary). Finally, I'm midway through lesson 1 of the DLI's Spanish Head Start course, which doesn't teach a ton, but what it covers it drills you on well. If you're a corporal, I know exactly what to say if we're introduced!

There's the background for our title: Fumbling Toward Polyglottery.

If you're going to be a polyglot, there's the question: How? Do you learn all your languages at once? Do you learn them all to fluency? Does it make more sense to learn them one at a time?

One of the challenges that comes up in becoming a polyglot is that you wind up a serial bilingual or serial trilingual instead. If you're going to be a true polyglot, you have to be able to have a bunch of languages running around in your head at the same time. So in a way, it's reassuring to see my learning go a little bit willy-nilly again. Keeping Spanish and Italian at the same time poses its particular set of challenges with respect to interference. I'm starting to discover, though, that if I study them both the same day, even during the same hour, instead of mixing them up, my brain catches on that I've switched. It's when I let one go dormant for a while that the interference starts, with me filling in the gaps in the weak language with guesses from the stronger one.

Want to be a polyglot? If you do, it's not just about learning multiple languages. It's about keeping them and letting them co-exist in your mind. So if you find yourself drifting between languages, instead of sticking to your schedule for learning them, don't worry. It's just practice for the day when you speak them all at a higher level.

Saturday, February 13, 2010

Michel Thomas - Can he still be your teacher?

I'm about halfway through the my Michel Thomas Italian review, and have noticed a few things. For one thing, he doesn't irritate me nearly as much as he used to. I remember the first time I did the Spanish course, his "push down on the PREsent tense" and "hit the ending" on the future about drove me batty. Ten years later, and with a sense that I still wasn't always stressing the right syllable, I understood what he was doing. I'm finding the same thing with the Italian course: It's the things that before I wished he'd let go and move on from that are still weak points for me - my brain knows but my mouth can't keep up - and I'm glad for the practice. Likewise, his tracks for verbs and other stuff like that, useful when you're learning, are fantastic when you're reviewing a language you'd just been mumbling along in for a few years. So in this sense, Michel Thomas is still a valuable teacher for me. But there's another sense too...

In my last post, on Assimil-ating a language, I mentioned that I was working through a Uighur text. I mentioned there that memory tricks are good for making language comprehensible in the short term, but that what I don't actually make use of I lose. I think this is part of the thing with Michel Thomas - the trick allows you to remember while he's teaching you, but it's the drilling that allows you to remember and automatically produce later. There's another thing I mentioned, though, that ties into this: Thomas' observation that what you understand, you know, and what you know you don't forget. I think there's more to this than I realized.

Since I've been studying Uighur, I decided to review my Uzbek a little bit. However, reviewing my Uzbek while doing a Michel Thomas course has me thinking in a different way than usual. Specifically, I've found myself asking, How would Michel Thomas teach this? It's given me some useful stuff. The present tense endings are almost identical to the pronouns; ditto for the "to be" suffixes:

I come : Men kelaman
I am American : Men Amerikalikman

But what's more, the past tense endings are very similar to the possessive suffixes:

My book : Kitobim
I came : Men keldim

So Michel Thomas might say that "You are what you do and you do what you are - the pronouns, the present tense endings and the endings for 'to be' are almost the same." But then, and this is the place where a thought popped into my brain and I thought Thomas had possessed me: "What you have done, you own, so to make the past, add a 'd' for done, then the possessive ending."

I don't know whether looking at the examples will be enough for it to make sense to readers who don't know anything about Turkic languages, but trust me, the way they go together it would be sweet to see Michel Thomas style courses for them. Unfortunately, Thomas is gone, his heirs haven't always measured up and Hodder and Stoughton isn't planning to make any more courses. But that doesn't mean you can't try to think of how your best teacher would have explained something. So if you're at odds with an idea or concept, stop trying to understand it for yourself. Try to understand it for someone else, imagining what your favorite teacher might have said to explain at least the part you get so far. You might be surprised at how things fall together.

Saturday, February 06, 2010

Assimil-ating non-Assimil materials

In my last post, I mentioned that I'd been playing around with an idea similar for getting similar benefits to the Assimil programs out of non-Assimil books. Today, I'd like to look at what I've been doing and how it works. This starts by considering just what it is that makes Assimil work, so let's start there.

Assimil books, as those who have used them know, work with facing page translations and extensive notes. The basic idea is that while working with the target language text, you can use your own language as a guide to what's going on. At a more advanced level, this can be problematic - the further you get into a language, the more you need to make it your own, not just a translation of a language you already speak. However, by the time people get to the advanced Assimil courses, they're mostly reading the target language text and glancing at their own language to double-check that they've understood, not to figure out what is being said. So what Assimil does is to give beginners a clear idea what's going on in the target language and advanced learners lots of good reading content. Cue the Michel Thomas quote:
What you understand, you know. And what you know, you do not forget.
There are lots of language textbooks out there that give you dialogs with translations. And there are others that eschew translations but have extensive notes, grammar explanations, etc. What happens in both cases, however, is that the learner has trouble assimilating the language because either it is unclear which words connect with which in the translations, or it is difficult to read the dialogs without turning it into a decoding exercise in flipping between the vocabulary, grammar explanations and the text.

What I have been doing with a Uighur textbook, Greetings from the Teklimakan, is to create a sort of passive phase Assimil course without an Assimil book. Here's what I do:

1) Learn the vocabulary list by whatever means necessary: link words, associations with words from other languages, etc. The technique is not important, because you only need to remember for five to ten minutes

2) Read the dialogs as best you can. Don't worry about understanding how everything goes together, about grammar rules, or whatever. Just try to make some basic sense of what's interacting with what.

3) Skim the grammar summaries, paying attention only to the sections with charts or endings. The idea is not to learn the endings, just to recognize what letter combinations are endings and whether they go with nouns, adjectives or verbs.

4) Re-read the dialog, seeing if you now recognize words you previously didn't since you've got a better idea what's the root and what's an ending.

5) Re-read the dialog once more and see if anything new falls into place.

6) Re-read the dialog the next day. If you think you mostly understand it, you can move on to the next one. If you've got doubts, repeat steps 1-5.

At no point should you actually try to learn anything. This is the passive phase. The idea, rather, is to get to where you're reading text in your target language and sort of understand what's going on.

A last thing: Every five lessons, you should probably skim the last five lessons and make sure they make at least as much sense as when you worked through them the first time.

I did the above with the Uyghur text for the first five lessons, including the re-read of all five lessons' dialogs. What I found is that my understanding of lesson 1-4 was actually pretty solid. Lesson 5 was weaker, but after a second reading, it started falling into place.

Note that if you want to, you can also put the words into Anki or somesuch. I did this, but I find that my recall is far better seeing the words in the context of the dialogs I know them from than in isolation.

When I have read all fifteen chapters this way, I will go back and work through the course - I haven't found a good version of Assimil's active phase to effectively apply yet. The nice thing about this system so far, though, is that I've built a passive vocabulary of around 300 words and I've learned a half-dozen structures without actively trying to learn any. And, I expect, working through the lessons when I'm done with this should be a breeze, since by then I'll be a false-beginner, not a neophyte.

The main thing we're looking at here, as is so often the case on this blog, is creating comprehensible input. New to a language and having trouble finding good materials for beginners? Apply this method to your materials as best you can and work a little ways into your text. Then, when you start to work in earnest, you'll have already assimilated hundreds of words of vocabulary and a few basic sentence types and you can get started with the "whys" of a language you already speak a little bit of, instead of everything being brand new.