Sunday, September 27, 2009

Natural Approach?

About a month ago, Tower of Confusion had a nice write-up on the Natural Approach and, for want of a better word, some Less Natural Approaches. At issue: Does a natural approach truly require "learning like a child"? The answer:
I believe that, as adults, we are smarter than kids, and we should be able to acquire languages more efficiently than them.
I think this is about right. Children have a lot of time on their hands with which to learn language - and lots else - in context. Adults have less time, but they have a lot more knowledge about the contexts they're learning about. As a result, most direct methods or natural approaches for adults teach things in sequences to take advantage of the fact that adults already know what sentences are and the types of things they express, etc.

There's a second question in here, and that's the introduction of artificial tools. I think there can definitely be an over-reliance on things like flashcards, grammar books, etc. On the other hand, if they work for you, it would be silly to eschew them. I think the key thing to keep in mind is that learning gets its start with comprehensible input. If you base your learning on the fact you've learned grammar rules or memorized a certain number of words, you're probably going to be disappointed when you first try to use your language for real. If, on the other hand, you're using artificial tools in order to make content comprehensible faster - eg, learning to recognize by endings which words are verbs or, for that matter, learning to recognize the 150 most common words in a language - that should be fine. Just so your consolidation of learning consists in meeting with content in a slightly less artificial form, there's nothing wrong with loading the dice in your favor as far as working with that content.

Friday, September 18, 2009

More Dead Languages!

In my search for information on old Celtic tongues, etc, I ran across a magnificent site:

Early Indo-European Online

This is a University of Texas resource including writings by the historical linguist, Winfred Lehmann, and more. The coolest part, however, is to be found on the sidebar of the page I've linked - lessons in 15 early Indo-European tongues. Of course the staples, Latin, Greek and Sanskrit, are included. And Old English and Old French aren't that exotic. But unless you want to spend serious money, good resources for Old Irish are few and far between (apart from Stifter's Intro to Old Irish and EV Gordon's Old Norse). And then there are those that only the most hardcore amateur historical linguist will have delved into: Gothic, Old Church Slavonic and, under the Old Iranian label, Avestan and Old Persian.

Typically, the lessons consist of an introduction to the language, then ten readings with detailed parsings. The grammar of the language is introduced step by step, across the ten lessons. Finally, there's a master glossary with every word form in the lessons and a second glossary of base forms that, by virtue of the links, will show you every usage of the word throughout the readings in all its forms. The second glossary also includes reference to the Indo-European source for selected words.

The lessons do fall a bit short of what I'd like to see for dead languages, in terms of treating them like modern languages and bringing them alive. And the site assumes that you know a little bit about how Indo-European languages work, what linguists are talking about when they start chattering about phonology, morphology, syntax and lexicon, and that you'll take the trouble to look up and learn more about, say, initial consonant mutations or verb-second languages on your own if these are unfamiliar. On the other hand, if you've got a little Latin or Greek, you should know enough to start wading in.

My favorite bit so far is from the Old Irish lessons (and you can tell how fast I'm wading by the fact this comes from lesson one): this image from the Compert Con Culainn, about a flock of birds:

Ba hálaind ocus ba caín in ténlorg ocus in ténarmhar boí leu.
Noí fichit én dóibh, rond argit eter cach dá én.
Cach fiche inna lurg fo leith, noí luirg dóibh.

Beautiful and delightful were the birds' flight and song,
Nine bands of twenty, and each two bound by silver chain.
Each band broke away in flight so there were nine trails.
(my own, doubtless inaccurate rendering)

And just in case this looks totally alien, here are a few words made more familiar:
noí, Latin novem, English nine
argit, Latin argentum, English silver
eter, Latin inter, between
dá, Latin duo, English two

If you're curious about languages of the past, this is a great site to visit, with the caveat that if you haven't done a language like Greek, Latin or Russian, you may want to start with Old English, Old French or Latin before tackling the more exotic offerings.

Saturday, September 12, 2009

Language Learning is Language Using

The Omniglot notes this story on how children learn language best: In conversing with adults. It's an important point for language learners - especially the self-taught - to keep in mind. Language is not a thing, existing in some perfect form and to be mastered by the memorization of its forms. Language is an action, or rather, a series of habitual actions subject to modification according to time and place for the purpose of achieving real world purposes.

If there were no such thing as a glass of milk, language is theoretically flexible enough to form a phrase to ask for one anyway, but such a phrase wouldn't get much use, and would quickly fall into desuetude (but I repeat myself). The same is true of the language you are learning through self-study: if it doesn't get anything suggestive of real-world use, it will at best crystallize as a sort of inanimate block of knowledge in your brain; at worst it will shrink away. If the only time you make use of your language is when you are studying, should you need to use that language you may well be stuck because you'll be in the wrong context for retrieving it (the newest studies say our memory problems are much like our real world organizational problems - it's not that the memory is lost but that you can't figure out the best way to retrace your mental pathways to figure out where you put it - see here). This means that if you want your language to come to you easily and naturally, you need to be doing things not just to learn but also to build the pathways so that it will be accessible when you want it.

Returning to the linked article, it suggests that kids who are read to a lot learn language, of course, but that those who actually converse with adults learn better. That's because language is a two-way street. It's not just for information reception; it's for communication, and you need to develop the capacity to formulate, not just receive thought.

I've written in the past about self-talk - learning key phrases for key situations and having a mental dialog with yourself to practice your language when you're in line at the bank or grocery store or whatever. To this I would add that it's useful to get yourself a small phrasebook so you can look up anything you've forgotten or haven't learned. By the third or fourth time you run through how to order what you want at the restaurant, ask for the check, etc, you won't need the phrasebook anymore - your brain will associate pulling up that language with going to a restaurant just as surely as you ask yourself if you shouldn't pick up an extra loaf of bread since you're at the grocery store anyway. The idea is that just as children get better at language by actually communicating - talking and not just listening - you can bring your language off the page and into your life more easily if you avoid associating it just with study time and mentally trick yourself into the belief that it's something you need and use in real-life situations.

Saturday, September 05, 2009

Re-ConLanging II

One of the trickiest parts of learning a new language is getting a hang of the pronunciation. Even with a language with an incredibly easy and consistent phonological system, like Spanish, you still have to learn to speak differently from in your own tongue. One of the bugbears of learning a dead or re-constructed language is that in the absence of native speakers, or even recordings of same, we don't know how the language was actually spoken. Comes the question: Should this stop us?

I am not an expert of Egyptology, but I have fiddled with a grammar or two. One of the things that comes up is the way that Egyptologists speak the language when muttering to themselves. It seems to be fairly consistent, from what I understand, and it seems to help them - it brings the language alive for it to be a spoken thing in the same way that learning written Mandarin would be even harder if you weren't associating it with the sounds of the tongue. So it would be neat if the Egyptian courses included a CD of somebody reading a few passages from the Book of the Dead, or at least a link to a site where you could download a few MP3s. (If it's out there, please let me know in the comments.) They say that you can't speak ancient Egyptian anymore because no one knows how it was spoken. True enough. But we can find out how it is spoken by a community of speakers who know it as a second or third language, at least. If you find yourself with access to texts in hieroglyphics and mutter your way through figuring them out, after all, do you want to sound like just another poor sap trying to sound out Budge's transcriptions from a Dover reprint, or do you want to sound like one of the leading Egyptologists of today?

This return to spoken courses comes not entirely out of the blue, of course. Of late, I've been fussing with Gaulish, and that led me, for comparison, to pull out Stifter's Intro to Old Irish. Stifter has a great system for helping you understand what's going on with the phonological changes in old Irish. His transcription system let's you see at a glance when lenition or palatalization has taken place, and gives guidance as to which word is causing the different initial mutations to unfold through the phrase. The problem is that working through a pronunciation guide focusing on individual sounds and working out how to fluidly pronounce ten or fifteen lines of text in succession are very different things. It would be neat to find at least a few of the poems on line, pronounced by a scholar, so that as a student I could better hear the changes that are made so clearly in the transcriptions. I'll google around for these, some more, and again, if anyone knows a good source, please note it in the comments.

Absent decent audio resources, of course, there's only one thing to do: Make them for yourself. And this is a lot easier than it used to be - plug a mic into your computer, then record and edit the sound files with Audacity and you can make a pretty well-ordered, clear-sounding recording. But it would be nice to have the guesswork about pronunciation taken care of by somebody who knows what they're doing :)

On a bright spot: Teach Yourself Beginner's Latin and Teach Yourself Old English both fairly decent audio. Assimil Latin has audio. You can get audio for a lot of the stuff associated with Wheelock. And now, from what I understand, Bolchazy-Carducci's latest Latin textbook series has MP3s available and configured to work conveniently with your iPod or Zune (though the program is expensive). So with Latin and Old English (!) available, maybe we'll be bringing still more dead languages to life soon.