Monday, April 26, 2010

Teach Yourself Complete...

This weekend, I picked up Teach Yourself Complete Italian. I'd been looking at the program for a while, and wondering if it would be better than the more recent Teach Yourself courses. So far, I'm moderately impressed.

In recent years, Teach Yourself courses have sat astride the past and present. They weren't as grammar focused as before, and the dialogs were meant to prepare you for real life. But the learning itself was only a step or two away from grammar-translation. What changed was the meaningfulness of what you were learning the old way, not how you learn.

With Teach Yourself Complete, it feels like some real strides have been made. There seems to be more emphasis on content and doing something with it and less emphasis on puzzling things out by means of vocabulary lists. What I like best, though, is that Teach Yourself Complete Italian feels like the kind of course that is meant to use CDs, not the kind for which they were thrown in as an afterthought. This means, in my view, that you shouldn't buy the text without the CDs. This point is made throughout the preface and introduction, by the way, which makes one wonder why they sell the text without the CDs at all.

If you take up the whole package - text plus CDs - and load the CDs on your iPod, you wind up with a fairly handy program for learning Italian 5-15 minutes at a time. If you're short of time, you can do one dialog. If you've got half an hour to kill, you can make it halfway through a chapter. In either case, you should start by listening, then turn to the book to make sure you've understood. In this way, for the first time, you can really use a Teach Yourself course to learn a language by ear while having a text as a fallback, instead of the other way around.

They're planning to roll out this course for Mandarin, Korean, Japanese and Russian in the fall. It looks as though Teach Yourself wants to offer a much cheaper but pretty good alternative to Living Language's Ultimate series. I'm just hoping it takes off enough that they'll upgrade some of their other exotics, starting with Turkish.

Sunday, April 18, 2010

Problems of Translation in Language Learning

When you listen to a Michel Thomas program, you spend a lot of time translating phrases from English to the language you are learning. And yet, when you're done, there's a lot that you can say comfortably, including phrases of your own devising. It occurred to me the other day why this should be so. It turns out that yet another aspect of the Michel Thomas programs that can get pretty irritating at first blush serves a purpose:

When Thomas introduces present tense verbs, he always makes a point of talking about how English uses "do" and "don't" with the verb or "am ...ing" and "is ...ing" to express things that other languages express with the simple present. For example, we say "Do you speak French?" The French just say, "Parlez-vous fran├žais?" We say, "I am leaving." The Spanish say, "Salgo." Thomas spends a fair amount of time trying to trick his students into making mistakes by attempting to use the English structures in the language they're learning, until they finally give up translating word by word and end up using the structures he's been drilling them to use.

It is a paradox, but by his constant harping on the errors that arise from literal translation, Thomas creates a course where the students spend 90% of their time translating phrases from English to the target language and end by being able to speak without translating. At least this is my experience - by the end of the course, the structures are ingrained, as are the meanings they imply in a more abstract sense. You end by knowing that in German, if you want someone to do something you start out "Wollen Sie..." and you talk about what you're going to do by starting with "ich werde..." and you cringe at the thought of considering the fact that the verb in English is "will" in both cases because you've been hit over the head so many times with the fact that if you think about what you're translating, not what you want to convey, you're going to get it wrong. By his use of situational translation and his direct addressing of the problems of literal translation, Thomas teaches you to use phrases and structures on auto-pilot when, in a target language only course you might well have recourse to translating in your own mind without realizing the problems you're creating for yourself.

The problem of translation, and the genius of Thomas' approach occur to me because of another curiosity I stumbled upon the other day: the phenomenon of staying in the target language when you speak but translating when you listen. This is something that was a problem for me in graduate school for French literature: I knew exactly what I was saying and I knew enough French to say it reasonably effectively (some of my professors may have disagreed on this last point). However, I did not have the same confidence that I understood precisely the point my professors were trying to make or that I fully grasped why an author had said something one way, rather than another. As a result, I subconsciously would be stopping myself every six or seven sentences to find an English equivalent of a distinction I couldn't quite make in French. I soon stopped because 1) it didn't work and 2) it was tiring. But I didn't take this decision of my own will; it's more that I gave in to the fact of imperfect understanding.

A lot of people talk about the importance of speaking a new language early, and of learning to think in the language. And one of the things that you hear is that you've got to get out there and speak, you've got to use the words you can and use the feedback you get. But with listening, it's different. People can't get in your head to know what you're thinking, so unless your answers are painfully slow or totally irrelevant to what was being discussed, they have no idea what's going on with your listening process? Do you even know? It's recently occurred to me that I can say lots of things in Uzbek, but I have a helluva time understanding an Uzbek song for which I haven't found the lyrics and done a translation. The problem? Even if I feel exactly what the line said, I can't think fast enough to translate it into English to make sure I "really understood."

I'm not sure how common this problem is, but I have met other people who do experience the same problem - thinking that they might not have understood something till they've translated it into English to make sure - and who therefore speak more easily than they listen. If you're having this problem, listen to the radio. Watch movies without subtitles. Do whatever you can to get used to listening to things you don't understand, and in lieu of getting transcripts or re-listening till you get it, try to develop an attitude that sometimes speech comes at you and all you can have is what you get right here and now. That way, when you're in a conversation with someone who talks too fast or above your level, you can listen for concepts, ideas and the direction of thought instead of getting bogged down trying to understand everything. In other words, don't sit around listening to the same thing twenty times, double-checking transcripts and asking follow-up questions till you're sure you know how to listen perfectly: Get out there and start listening to your target language now!

Sunday, April 11, 2010

A Michel Thomas forerunner?

I recently came across An Arabic Primer by Sir Arthur Cotton. His premise is quite interesting:
Languages are usually learnt as if it took a long time to learn the grammar &c., but that to speak with a good pronunciation and expression, and freely, and to catch the words from a speaker by the ear were easily and quickly acquired, but this is exactly contrary to fact.
As a result, he proposes to teach Arabic by giving you 30 key words, then learning 180 sentences from those words.

Now, the key element of Michel Thomas - instructor focused methodology - is missing, but the idea of breaking down the language to simple elements and using a limited vocabulary to develop a facility for combining those elements is quite plainly there. Skimming the book, it looks like you'd finish with the ability to make a couple 100 sentences and to make your own sentences by learning words on your own and adding them to the structures. The means by which this is accomplished is to take a native speaker and have him repeat for you, over and over, while you repeat for him in turn, always without moving onto the next thing till the last thing has been mastered.

Arthur Cotton's Arabic primer is, of course, far behind the best of today's courses, though it might still be ahead of some of the worst. But for a language program that's over 100 years old, it looks pretty well conceived. If you're interested in trying it, it's out there. And even better, you don't need to find an Arabic speaker to use it. Librivox has a recording where the reader takes you through the process, so all you need is an iPod and lots of time and patience. You can find it here.

Sunday, April 04, 2010

Fun Links

The last week or so, I've been working on two things: filling in gaps (slowly) in my Uzbek vocabulary, and taking another look at the Romance languages. I've got a sense of some basic Uzbek structures and what to do with them, but words are wanting so I decided it was time to be able to talk about some new things. I've got some things I've been trying for this. If any of them work out, I'll update on that next week.

Reviewing the Romance languages has been inspired by two sites, one with some excellent input - www.bilingual-texts.com - and one that proposes a modern Latin as a tool to getting in touch with the major Romance languages - http://sites.google.com/site/neolatino/.

Bilingual-Texts.com is an excellent site with a small (but growing) selection of, well, bilingual texts. You can get selected novels and play in French, Italian, Spanish, Russian and Japanese (maybe more by now) with English translations. And in those cases where audio is available free online, this is linked. If you're an advanced beginner or higher and what some good material with the roadmap provided by an English translation, be sure to check out the site. Or, if you're a purist, you can cover the second column and still enjoy the material!

When it comes to the neolatino site, I'm a bit of a skeptic. Neolatino is an auxiliary language based on Latin, and I'm a skeptic of auxlangs in general. However, if you can read a couple Romance languages, it's fun to poke around, to see how easily one can read this "neolatino" and to see, equally, how little effort it takes to read texts in languages like Valenciana, once you get out of the habit of reading for specific meaning in a language you know and start reading for general meaning as best you can. If you really should be studying your French/Spanish/Italian/Portuguese, etc, but you don't feel like it, this is a great site to kill some time.