1317 messages over 165 pages: << Previous 1 2 3 4 5 6 7 ... 78 ... 164 165 Next >>
csidler Diglot Pro Member Australia chadsidler.com Joined 4832 days ago 51 posts - 59 votes Speaks: English*, German Studies: Italian, French Personal Language Map
| Message 617 of 1317 11 July 2013 at 9:23pm | IP Logged |
Wow.
I have spent the past few weeks reading this entire log from beginning to "end", or should I say, to this page,
because there is no end!
I have enjoyed it so far, however wasn't happy when you switched entirely to French for a bit, but luckily for
me you eventually came back to English!
You have been quite inspiring, and providing great insights and revelations, with the highlighted point being
that success requires hard, persistent, and consistent effort. Also that if you want to be able to do something,
then you need to just do it!
I look forward to beginning my own French journey soon, as well as applying your inspiration to my German,
and I look forward to reading your log in "real time", now that I have ploughed through everything and am up
to date!
1 person has voted this message useful
|
emk Diglot Moderator United States Joined 5541 days ago 2615 posts - 8806 votes Speaks: English*, FrenchB2 Studies: Spanish, Ancient Egyptian Personal Language Map
| Message 618 of 1317 11 July 2013 at 9:32pm | IP Logged |
As some folks might have noticed, this log contains a mix of victories and occasional frustrations. Language learning is a bumpy process, with high points and low points. I try to write about both, because I want to capture the process. And sometimes it helps to share our frustrations.
I don't usually make a long list of "reasons why language X is hard", because that's normally not how it feels to me. French can be a little weird at times, but it's really one of the easier languages for an English speaker, and in any case, enough exposure and native materials will make just about anything seem normal. All that stuff that bugs first-year French students? I've seen it tens of thousands of times, so it feels pretty ordinary and straightforward to me.
But today I'm going to kvetch a bit, just to get it out of my system. Feel free to skip this.
...
Ah, gender. Let's talk about gender.
Gender in French is tricky. On the bright side, French has only two genders, not three or fifteen, and there aren't any cases. But after that, there are complications:
1. The gender of words is only about 75% predictable from the word endings, but the pattern is sufficiently complicated that it takes a book to lay it out. It's not like Egyptian, where all the feminine nouns end in "-t", or Italian, where the word endings are a lot simpler. However, French kids do rely on word endings, at least in part, up to about age 5, at which point a more-developed gender system apparently starts to take over:
Quote:
Dans une autre recherche, Karmiloff-Smith (1979, Exp. 9) introduit un rapport conflictuel entre le genre véhiculé par la terminaison du mot et celui de l'article (ex. " Voici l'image d'une-F bicron-M "). L'enfant doit de nouveau décrire une transformation effectuée sur l'image. Les jeunes enfants en-dessous de cinq ans sont très ambivalents dans leurs choix : dans près de la moitié des cas, le genre attribué est identique à celui du suffixe (ex. " Vous avez caché la bicron vert "). A partir de six ans seulement, l'information de genre sur l'article semble l'emporter sur la marque comme indice du genre (ex. " Vous avez caché la bicron verte "). Les commentaires des enfants suggèrent qu'ils sont particulièrement gênés par l'incompatibilité des deux informations. Pour contourner cette difficulté, ils mettent en place deux stratégies : l'ellipse du substantif (ex. " la grise " au lieu de " la bicron grise ") et l'alternance de la marque (ex. " la bicronne "). |
|
|
And even French adults will mess up less common words with misleading endings. And then there's this really weird study, where native speakers did surprisingly poorly.
2. Vowels at the start of words cause problems, because they hide a lot of gender information. There's no way to tell whether l'eau (the water) is masculine or feminine, because you can't see the vowel in the article.
3. Vowels at the start of words have strange effects on adjective selection, too. You say un beau mec (a handsome guy, "a hunk"), un bel homme (a handsome man) and une belle femme (a beautiful woman). The masculine adjective bel sounds just like the feminine adjective belle, and you use it because homme starts with a vowel sound in French. There's actually a research paper somewhere that claims this kind of adjective selection requires a different neural pipeline than that used by German speakers, because more work needs to be delayed until the phonetic form of the noun is available. And all this interacts with liaison, too.
4. Lots of important gender information is encoded using French's rich vowel system, which means that non-native speakers may literally fail to hear it if they don't make an effort. It's important to distinguish between the vowels in -on and -onne, -en and -enne, and so on. In many cases, this interacts with the stuff mentioned in (3). Some English speakers even get sloppy about le and la, smearing the vowels into a mush and hoping nobody will notice.
Unsurprisingly, French gender can be challenging for anglophones. In fact, it seems to fall into the same category as accent: Adult learners can do very well, but they'll still make mistakes. There's the notorious case of the actress Jane Birken, an English speaker who moved to France at age 14. Over fifty years later, she speaks very reasonable French, but she's well known for making mistakes with gender. Of course, many French people suspect Birken may be doing it deliberately at this point, and some other adult learners do considerably better.
So French gender is complicated. French children finish acquiring it relatively late, French adults can still mess it up in certain circumstances, and a few adult learners still get it wrong after 50 years of immersion.
To be fair, none of this matters very much if you're only aiming for B2 or so. It's not too hard to do a reasonable job with French gender, especially if you speak at the lower end of ordinary conversational speed. But as the last week of immersion has once again shown me, life's a lot easier when you can speak fast. And personally, my ultimate goals are well above B2.
So what I really want is a native-like "gender system": I want to learn and use gender the way a native does, picking up word genders unconsciously and using them correctly in rapid speech without having to think about them.
How acquiring a language feature works for me
I've internalized three aspects of French which have little relationship to English: the clitic pronouns y and en, and the subjunctive. In each case, my learning process went through a number of stages:
1. First I'd learn the "theory", and wait for quite a while.
2. One day, I'd start to notice the feature very clearly when I heard it.
3. A while later, I'd feel a pronoun like en "hovering" over sentence. My brain wanted to stick it in somewhere, but wasn't sure where.
4. I'd eventually start to use the feature mostly correctly, with some mistakes.
5. The mistakes would gradually go away.
My progress with gender
I already have a dodgy "gender system," which is good enough to get articles and many adjectives right for common nouns. But what I want is what the French six-year-olds demonstrated in the Karmiloff-Smith study mentioned above: A gender system which automatically picks up on weird exceptions, and which can handle long-range agreements tolerably well.
I've spent much of the past year working on two things: (1) forcing myself to notice gender markers in input, and (2) trying very hard to get gender right in output, with no bluffing.
This has really paid off in a lot of ways. I do notice a lot of gender markers in input. When novice French students say things like un voiture, it sounds blatantly wrong to my ears. When a native speaker uses a gender I don't expect, I notice it—and I get that word wrong less often in the future. And if I miss a word, I can sometimes use the gender of the surrounding words to guess it. So I'm at roughly stage (2) on my scale above.
But crossing over into stage (3) is rocky. My brain knows that it needs to use adjectives and articles with the correct gender, but some of the details are still hazy. So when I talk too quickly to patch up my speech consciously, my brain is inserting gender markers mixed results. And my French vowel system is getting better, too, which makes more gender markers visible.
And as I mentioned, I'm the sole English-speaking adult in a group of French speakers this week, and I need to speak fast. And thanks to my partial progress with gender, some sentences feel like minefields of potential mistakes these days.
This would all be easier if I slowed down a bit, and took the time to patch up gender manually. I can do that tolerably well. But I just don't have the right circuitry in my brain to get all this correct at high speed without thinking.
But there are bright moments, such as waking up and thinking les photos que j'ai prises even before my brain finishes coming on line.
Things I still don't understand
What are the learning tradeoffs with input (which I do), speaking (which I do), and drills (which I find tedious)? What are the limitations of mostly speaking with the same person?
Why can I speak comfortably some times, and trip all over my tongue at other times?
Has anybody else noticed that Khatzumoto's prescription for awkward speaking skills is to develop basically C2 listening comprehension? This seems rather unusual, to say the least.
Khatzumoto wrote:
Here is my take on speaking.
You said you’re listening is strong, and I’m sure it is. But how strong? Can you follow Trick 100%? Can you follow the Japanese Diet proceedings (www.shugiintv.go.jp) 100%? Can you follow Tiger and Dragon 100%? Can you repeat virtually any 5-15-second-long piece of dialogue you hear, verbatim, after one listening? If not, then, I’m going to go with the input hypothesis here and say that you do still need to listen EVEN MORE. |
|
|
And why, after several millennia of written history, are the upper levels of language learning such a mysterious process?
Oh, well. I may not have the answers, but I do have lots of cool French TV to watch. And I'm pleased to say that latter episodes of Le trône de fer are a bit more challenging than the first—not overwhelming, but I need pay attention and occasionally rewind some. This is good; I didn't want a super-easy series.
Edited by emk on 12 July 2013 at 1:20am
9 persons have voted this message useful
| DaraghM Diglot Senior Member Ireland Joined 6160 days ago 1947 posts - 2923 votes Speaks: English*, Spanish Studies: French, Russian, Hungarian
| Message 619 of 1317 12 July 2013 at 11:15am | IP Logged |
I remember you mentioning a book by Saul Rosenthal in your log, "French Key Words and Expressions". I picked up this book, along with a really good book he wrote on French gender,
The Rules for the Gender of French Nouns
I'd strongly recommend this book for French learners.
Edited by DaraghM on 12 July 2013 at 11:16am
2 persons have voted this message useful
|
emk Diglot Moderator United States Joined 5541 days ago 2615 posts - 8806 votes Speaks: English*, FrenchB2 Studies: Spanish, Ancient Egyptian Personal Language Map
| Message 620 of 1317 12 July 2013 at 6:05pm | IP Logged |
csidler wrote:
You have been quite inspiring, and providing great insights and revelations, with the highlighted point being
that success requires hard, persistent, and consistent effort. Also that if you want to be able to do something,
then you need to just do it! |
|
|
Thank you for your kind words! Although I should be honest—I might be fairly "persistent" and "consistent," but I don't necessarily work all that hard, as such. My log is full of Assimil and BDs and TV and science fiction novels because that's a huge part of what I do: I find something interesting that I can more-or-less understand, and I try to give it my undivided attention for a while. Even sentences and cloze-deletions in Anki are a pretty low-stress way to learn things. Speaking does sometimes require real work, but with any luck, that can be fun too.
I think it's a mistake to rely too much on sheer willpower and hard work. The voyage is just too damn long. The people who succeed either burn their ships and leave themselves no choice, or they find a way to enjoy the learning process itself. The latter gets easier once you can access native materials, because the world is full of incredibly addictive entertainment. But it's possible to find some fun in Assimil, and a few other beginner courses, too.
DaraghM wrote:
I remember you mentioning a book by Saul Rosenthal in your log, "French Key Words and Expressions". I picked up this book, along with a really good book he wrote on French gender,
The Rules for the Gender of French Nouns
I'd strongly recommend this book for French learners. |
|
|
Rosenthal's books are excellent, and I love his book on gender. There are also some free sites with similar information. I think that many French students, especially anglophones, would benefit from studying these rules. As I mentioned above, there are some scientific studies suggesting that French preschoolers make heavy use of word endings when they're first "bootstrapping" their gender system. This gives them a framework that they can later use to internalize the real rules. Native French teachers don't generally encourage students to use word endings, but I found it really useful.
My lesson today
Wow, today was so much better than a week ago. I could actually talk! Roughly speaking, my level last week was about halfway between B1 and B2, which is catastrophically bad for me. This week, I was back up between B2 and C1 where I belong.
My speaking is still too variable and fragile. My tutor suspects that some of this is due to fatigue, and some of it is because I've gone from A2 to a very solid B2 in a year and a half, and my French hasn't fully "settled in" yet. Personally, I also suspect that my brain is trying to perform several upgrades: it's installing a better gender system, adding two or three round front vowels, and trying to sort out some gender-related issues with my nasal vowels. Plus, as always, there are thousands of words and turns of phrase hanging out just below the "activation threshold." All of this ongoing "construction work" can really hurt me on a bad day.
But the lesson was great. We spent a lot of time talking about sales copy, and why you really need to see your offering through your customer's eyes. What do (paying) language students actually want? Some want to fit into a new society, others want to succeed at work, some want to attend a university and therefore need to pass an exam. They want to know that their tutor can help them reach their goals. And if you want to talk to them, you should talk to them in terms of what they need.
Yeah, I'm basically paying my tutor for the privilege of giving her business advice. :-) It's half business role-play, and half real consulting, and it's really helping me to improve my business French. By discussing common topics in depth, I'm filling in vocabulary holes and building a repertoire of canned phrases. This works a lot like the "islands" technique in Shekhtman's How to Improve Your Foreign Language Immediately, except it's a lot more organic.
Role playing is great. So is using a second language to teach a skill. And now that I think about it, this offers some real possibility for language exchanges. You don't always have to teach people your native language. You could teach them photography, or cooking, or bookkeeping, or kayaking.
5 persons have voted this message useful
| Travis.H Triglot Groupie United States Joined 4465 days ago 59 posts - 91 votes Speaks: English*, Japanese, Sign Language Studies: French
| Message 621 of 1317 13 July 2013 at 2:25am | IP Logged |
One thing I've always wondered about Khatz is the 10,000 hour rule about listening and
its effect on speaking. The 10,000 hours idea coming from Gladwell's book about 10,000
hours being the baseline for mastering something. But what hasn't made sense to me is
where and how the 10,000 hours thing is defined. Is it 10,000 hours or 10,000 times?
Do I need to say a word 10,000 times to become a master at that word? Do I need to say
a sentence, or use a grammatical pattern 10,000 times to become a master at that one
pattern? With something as difficult to define as fluency or mastery in a language, it
seems even more difficult to determine where and how those 10,000 hours should be
spent.
In my personal experience, I've found that hanging out and talking with a group of
native speakers to be the most helpful in terms of developing more native-like fluency.
With Japanese, I'd listen and pay careful attention to how other members in the group
would insert their opinions and thoughts to ideas and then mimic that soon after,
attempting to maintain flow. Where I feel like I learned the most was when I would
explain something and someone would laugh at me for saying something incorrectly.
Other members in the group would typically chime in and I would get some sort of
correction. Somehow (and maybe this is because Japanese culture is group oriented)
having more people there encouraged the group to maintain normal conversation which
included correcting any strange speech. One-on-one conversation seems to let more
mistakes slide for the sake of communication continuing. Groups don't tend to have as
much patience as one-on-one does.
Strong listening skills help you develop more than your vocabulary. I believe that
well developed listening skills will help you intuitively know where verbal pauses are
supposed to go, and the timing. Often fluent conversation is as much about what is not
spoken as what is.
Edited by Travis.H on 13 July 2013 at 2:27am
1 person has voted this message useful
|
emk Diglot Moderator United States Joined 5541 days ago 2615 posts - 8806 votes Speaks: English*, FrenchB2 Studies: Spanish, Ancient Egyptian Personal Language Map
| Message 622 of 1317 13 July 2013 at 6:50pm | IP Logged |
10,000 hours
Travis.H wrote:
The 10,000 hours idea coming from Gladwell's book about 10,000
hours being the baseline for mastering something. But what hasn't made sense to me is
where and how the 10,000 hours thing is defined. Is it 10,000 hours or 10,000 times?
Do I need to say a word 10,000 times to become a master at that word? |
|
|
Let's just ignore Gladwell for minute, and take a look at the work of K. Anders Ericsson, who proposed the original "10,000 hours" rule long before Gladwell popularized it. Ericsson appears to be one of those scientists like Krashen: He has a single, really important idea that he defends well, but he doesn't do nuance. You can find a recent summary of his ideas on his website.
As far as I can tell, there are two important things to understand about the 10,000-hour rule:
1. 10,000 hours is a really long time. In the US, people work for about 2,000 hours per year, and maybe half that time is really productive if they're lucky. So 10,000 good hours is about a decade of your productive life, under ordinary circumstances. (To be fair, AJATT is not "ordinary circumstances.")
2. Ericsson's rule only applies to something called "deliberate practice". This does not mean that 10,000 hours of chess playing will make you a grand master, or that 10,000 hours of violin will make a word-class musician. It means 10,000 hours of hard work, outside of your comfort zone, specifically tackling your particular problems.
There are some researchers who disagree with Ericsson. You can find a good summary here. In short: Some people become expert in 3,000 hours. Other people are merely good after 10,000 hours. (Ericsson would say that these people have been engaging in ineffective practice, not "deliberate practice".) But even most of the researchers who disagree with Ericsson end up admitting that massive amounts of practice plays a huge role in developing expertise. And Ericsson observes that many people reach a comfortable level and give up on deliberate practice.
So how does this relate to language learning? Well, for one thing, it gives you an idea of just how hard-core native speakers really are. A native speaker who does well in school:
1) Has heard an average of 13 million spoken words per year since birth.
2) Has read over a million words per year (roughly half a Super Challenge) starting sometime around age 10.
3) Has spent thousands of hours using their language to persuade, to play, to charm, to learn, and so on.
In other words, pretty much everybody is an Olympic athlete or a Grand Master when it comes to their own language. But there's some good news for learners, too:
1) The first 20% of the effort goes a long way, like always.
2) An English speaker learning German or French gets everything at a 75% discount, if you believe the FSI numbers.
So, very roughly speaking: 10,000 hours * 20% * (100% - 75%) = 500 hours to reach a pretty useful level in a European language. I think most people reach B1+ in this time, though the FSI training programs use somewhat brutal methods and get people to ILR 3 (allegedly about C1) with 600 hours of class time and around 480 hours of homework.
Adult language acquisition
Scientific papers about adult language acquisition are really depressing. Many researchers seem to start out with one or more of the following assumptions (which I'm going to caricature shamelessly):
1) "The only interesting question is what it takes to become a flawless native speaker with a perfect accent. Somebody like Linus Torvalds is an example of failed language acquisition because he said 'one of the worst troublespot' instead of 'troublespots', and besides, a few of his vowels were weird, and no native speaker would ever do that"
2) "Children are magically gifted at learning languages, and succeed 100% of the time, but if adults do it, it's some kind of weird outlier that doesn't really count, possibly because it was reported by my gullible colleague who can't accurately evaluate speaking skills."
3) "Language acquisition in only interesting insofar as it supports my pet theories about the Universal Grammar, the Language Acquisition Device, etc., so I might as well look at everything through that lens."
Now, I'm being deliberately unfair here, because it's fun, and because I'm blowing off steam after reading way too many papers that don't control properly for total quantity of comprehensible exposure or for the student's level of integrative motivation. Or in colloquial language, "I don't think their research on adult learners counts unless they find adult learners with 100+ million words of spoken input and 10+ million words of written input, and those learners want to belong to their new culture with the same desperate craving that the average child feels."
If you give the adults the same chance that native kids get, they do pretty well. You can take somebody like Jane Birkin (a classic example of a teenage leaner who never became a "native" French speaker), and you know, she still does a pretty darn good job.
An interesting paper about adult language acquisition
Anyway, the other day I read a much better than average paper on adult language acquisition that I saw over on Leaky Grammar. This starts out from the assumption that adult language acquisition is both necessary and difficult, and it discusses various strategies used to cope with this challenge. There are some pretty realistic observations:
Quote:
We might then ask what proficiency would be required to serve communication between two groups speaking different languages in such an oral world. The following is a characterization of proficiency at the 2+ level taken from the Internet:
Quote:
Level 2+. Able to satisfy most work requirements with language usage that is often, but not always, accessible and effective. The individual shows considerable ability to communicate effectively on topics relating to particular interests and special fields of competence. Often shows a high degree of fluency and ease of speech, yet when under tension or pressure, the ability to use language effectively may deteriorate. Comprehension of normal native speech is typically nearly complete. The individual may miss cultural and local references and may require a native speaker to adjust to his/her limitations in some ways. Native speakers often perceive the individual's speech to contain awkward or inaccurate phrasing of ideas, mistaken time, space, and personal references, or to be in some way inappropriate, if not strictly incorrect. |
|
|
Proficiency in the 2 range would have been adequate for most intergroup communication and could be achieved by a talented learner in two years with sufficient exposure and use. We can imagine that target language-like discourse and grammar may not have been necessary as long as the individual was communicatively effective. |
|
|
But most insightful part is hidden in an afterthought:
Quote:
The story gets complicated though when we observe that many adult learners of a second language acquire high proficiency in the L2, and not infrequently immigrant children being raised in bilingual environments (typically children who are second or later in birth order) fail to acquire their heritage language and become essentially monolingual speakers of the language spoken in the wider community. |
|
|
I've seen both cases with my own eyes: Adults who reach near-native levels, and children who develop purely passive skills even though they have significant and constant exposure.
Age may influence accent and possibly the fine details of grammar. But the real deciding factors lie elsewhere, because adults succeed all the time, and children are only guaranteed to succeed under very specific circumstances.
5 persons have voted this message useful
| Cavesa Triglot Senior Member Czech Republic Joined 5018 days ago 3277 posts - 6779 votes Speaks: Czech*, FrenchC2, EnglishC1 Studies: Spanish, German, Italian
| Message 623 of 1317 13 July 2013 at 8:30pm | IP Logged |
I love your informative posts, emk. I'll look at the paper, it seems interesting.
I think even the best attempts on quantifying how many hours of exposure are needed (outside the comfort zone, suited to your needs etc.) are doomed to fail. Firstly, there is not such a vast amount of guinea pigs. People who study to become translators/interpreters are not usual language learners and most others are content to stop around B1 (or the lowest level for immigrant exam). And those who have succeeded usually don't have a detailed log on exactly how many hours they spent on the language over the years. And even if you got enough guinea pigs, there are few as varied and individual things as the human brain. There are too many kinds of learners from the point of view of intelligence, motivation, ways to learn, talents for this or that part of learning, conditions for learning, etc. and all that together. And there are differences depending on the target language.
But the 10 000 is a really beautiful number and a nice goal to strive for. If you are awesome after "only" 3000 hours, good for you! If you are not that awesome after the full 10000, than I still suppose you moved a lot forward and that is great as well. And all this is just another reason why it pays off to have fun while learning. Even if you end up far bellow the happy ending you dreamed off, you'll still be unlikely to regret the time.
2 persons have voted this message useful
| s_allard Triglot Senior Member Canada Joined 5439 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 624 of 1317 15 July 2013 at 10:19pm | IP Logged |
I think this 10,000 hour rule is more a pithy statement to sell books rather than a hard fact. But the point is that
learning anything takes time. That we know. What we are all trying to do is speed up the process and learn
effectively and efficiently.
I would like to briefly revisit the topic of the acquisition of the grammatical gender system in French. In my
opinion, this feature of French grammar is one of a number of differential features of French and English
grammar. The basic reason why certaiin things are difficult in a target language is that they do not exist in our
native language. For example, in French, in addition to grammatical gender, pronominal verbs are probably one
of the most difficult things to acquire.
Insofar as grammatical gender is concerned, it should be pointed out that the fundamental problem is not
learning the gender of nouns but how to work the gender morphosyntax. In other words, it's the system of
agreement that is difficult.
What that means is that any learning strategy of gender has to emphasize what I call morphosyntax linearity. For
example, the question isn't really whether véhicule is masculine or feminine. It's more a qusetion of whether you
say un véhicule neuf or une véhicule neuve. (The former is correct).
The key in learning the system is therefore to learn entire phrases that highlight the gender agreement system.
And let me end by mentioning the virtues, once again, of the site Fluent
French Now that @emk knows well. This is a site for serious students of French and not a shill for some
dodgy product.
Edited by s_allard on 16 July 2013 at 12:24am
2 persons have voted this message useful
|
You cannot post new topics in this forum - You cannot reply to topics in this forum - You cannot delete your posts in this forum You cannot edit your posts in this forum - You cannot create polls in this forum - You cannot vote in polls in this forum
This page was generated in 0.4688 seconds.
DHTML Menu By Milonic JavaScript
|