Register  Login  Active Topics  Maps  

The Cheating & Consolidating Method

 Language Learning Forum : Learning Techniques, Methods & Strategies Post Reply
136 messages over 17 pages: 1 2 35 6 7 ... 4 ... 16 17 Next >>
Serpent
Octoglot
Senior Member
Russian Federation
serpent-849.livejour
Joined 4671 days ago

9753 posts - 15775 votes 
4 sounds
Speaks: Russian*, English, FinnishC1, Latin, German, Italian, Spanish, Portuguese
Studies: Danish, Romanian, Polish, Belarusian, Ukrainian, Croatian, Slovenian, Catalan, Czech, Galician, Dutch, Swedish

 
 Message 25 of 136
02 August 2014 at 12:31am | IP Logged 
Expugnator, I think GLOSS is a good tool to use in your situation.

@s_allard, imo if you don't know the meaning in a given context, for this particular text/paragraph the word counts as unknown, even if it's not unfamiliar.
1 person has voted this message useful





emk
Diglot
Moderator
United States
Joined 3606 days ago

2615 posts - 8805 votes 
Speaks: English*, FrenchB2
Studies: Spanish, Ancient Egyptian
Personal Language Map

 
 Message 26 of 136
02 August 2014 at 1:23am | IP Logged 
s_allard wrote:
emk wrote:
Or How I Reached 99.8% Comprehension of (Some) French Fiction While Goofing Off
...

I'm assuming that there is tongue in cheek here. From what I read in the rest of the post, the effort involved in
reaching that level of comprehension of French hardly qualifies as goofing off. That was some serious work!

Well, it was certainly quite a few hours—probably over 400 hours for extensive reading, around 90 hours for Anki reviews, and some more time spent creating Anki cards, all over the course of about 30 months. But at the same time, I wouldn't call it "work"—most of that time was spent reading science fiction novels, random French websites and bandes dessinées, or watching TV series on DVD. I never did any really strenuous vocabulary work, just easy sentence cards.

YnEoS wrote:
So at the moment I don't really have a good system of when I should focus on consolidating or when I should be "cheating" to make new words decipherable. Usually I just do an activity that seems to help until it stops helping, then I find some other activity and through lots of awkward trial and error knowledge and progress eventually happens.

That's pretty much what I do, too. At first, it's hard to much more than slowly decipher simple texts, and the easiest way to consolidate that knowledge is to re-read the same texts a bunch. But if you look at the early days of my log, I totally just plowed through interesting books where 10—20% of the content was hopelessly opaque, and much of the rest involved a large element of guesswork. But then again, this works better in related languages with lots of cognates.

As soon as I could read things for fun (even with imperfect comprehension), I personally enjoyed spending a lot of time consolidating, and I tried to streamline dictionaries, etc., as much as possible. For me, one of the most powerful forms of "cheating" is being caught up so deeply in a work that I can't put it down. Being totally engrossed with a work for hours at a time gives me a huge comprehension boost. Again, this is not a new hypothesis: Krashen calls it "compelling input."

But you know, I don't worry too much over proportions. If I feel like vocab is an annoying obstacle, I spend some time on it (or find a more creative way to cheat). If I want to read a fun book without a dictionary, then I go for it. If I want to mix up different activities, I do.

Where people really get into trouble is when they either (1) get too little comprehensible input to ever really consolidate anything, or (2) they try to consolidate with mostly opaque materials. Another way of saying this is, "I think it matters to get lots and lots of 'i+1' input." That's the sweet spot. But every learner, every language and every stage of the process will probably require tweaking and adjustment.

Expugnator wrote:
I'm
constantly learning at least 1 'transparent' language and 1 'non-transparent' language
at the same time and I'm pretty much sure they're two different beasts. You may learn
enough to conversate comfortably in the non-transparent one, but deciphering a native
text requires learning a whole lot of words you took for granted when learning another
Romance or Germanic language.

I also have a Romance language and an opaque language, and yeah, they do feel different. I talk more about my Egyptian process here: I'm basically mining Assimil for cloze cards. It's going pretty well.

For a while, I was also doing some extensive reading in Egyptian. I was using interlinear texts from the St. Andrews corpus, particular this edition of the Westcar papyrus. My process was a bit weird: I'd try read all three versions of the text (hieroglyphs, translation and English), and try to figure out what corresponded to what. If I couldn't, well, no worries. Then I would finish by reading the Egyptian version using my newly "boosted" comprehension, and I would actually try to understand it directly as Egyptian. This is a bit hard to describe—it's sort of like watching a scene several times with subtitles, but then watching it one final time with just the L2 audio.

This was actually working pretty well, despite my low level. I've put it on hold to work through Assimil (and to handle professional stuff), but I'll try it again in a while. Of course, one of the big barriers with Egyptian is that it's hard to find thousands of pages of interesting material, and sometimes nobody has a good grasp on the intellectual or bureaucratic vocabulary. But so far, things are working reasonably well.

garyb wrote:
So I'm mostly just thinking aloud here, and sorry if I'm going off-topic a bit since the topic is about passive skills, but I'm wondering if the "cheating and consolidating" principle can also be applied to active skills.

I suspect this is true, to some extent. One thing I notice is that my passive knowledge is basically a database that I can call on consciously, and manually assemble into coherent French. My passive knowledge is basically a large, poorly-consolidated cheat-sheet for active production.

And to a certain extend, I've consolidated all my "day to day" output. But once I get out of my comfort zone, it's a lot more hit or miss. What I would really love to try would be living, studying and working professionally in French for three months to a year.

But it's worth keeping in mind that I do lots of other stuff that isn't in this post, too, especially on the output side. This post focuses on the input side.

s_allard wrote:
But I do want to raise the issue of comprehension, especially when we attach a figure such as 50%, 70% and even 99.8%.

The 99.8% number is based on the last several ebooks I've read. I have a habit of highlighting sentences which contain unknown vocabulary or interesting new uses of known vocabulary. So basically, in this context, "Comprehensible" counts both "Automatic" and easily "Decipherable" vocabulary. It's not a terribly precise count. In my most recent book, I was averaging one marked word every four pages or so. Assuming 350 words per page (typical for mass-market paperbacks), that works out to about 1 difficult word in every 1400.

Every once in a while I'll run into a text with 1 opaque word per 100. These are often very high-brow political editorials, or extremely colloquial websites aimed at audiences in their teens and twenties.

But in general, vocabulary counts are tricky, and vary considerably depending on methodology. I try choose a reasonable methodology, and then to explain how I applied it.

luke wrote:
I think it's a great framework for creative ideas as well as seeing how well we can all shoehorn various techniques into it.

This is pretty much my goal with this post: creating a framework that makes it easier to discover fun new ways to cheat, particularly in areas like listening comprehension.

4 persons have voted this message useful



s_allard
Triglot
Senior Member
Canada
Joined 3504 days ago

2704 posts - 5424 votes 
Speaks: French*, English, Spanish
Studies: Polish

 
 Message 27 of 136
02 August 2014 at 2:54am | IP Logged 
Measuring comprehension is not the same as counting unknown words or unknown meanings in a text. The
methodological problem is that it is difficult to determine if one has really understood the words on the page,
especially with works of fiction where the enjoyment and appreciation is determined by things such as style and
story. We see this in works such as comic books and movies or tv where there are visual and auditory clues that
add context and meaning to the words. As I alluded to earlier, one often does not need to understand or even
hear all the words to appreciate some works.

Works of fiction often tend to be rich in idioms and stylistic devices such as metaphors, synecdoche, simile,
irony, sarcasm, etc. This is why humour, especially standup comics, is difficult to understand.

Because of these issues, language proficiency tests such as the CEFR prefer non-fiction texts where questions
about specific subjects and facts can be readily constructed. And even there, it is often quite astonishing how
often people can differ in their understanding. In my classes, I'll ask if everyone has understood a certain text.
Everybody says yes. Then I ask people to tell me what they understood. I'm always surprised how people can read
a single page of non-fiction and have completely different understandings. With fiction, I wouldn't be surprised if
there were even more variation.




1 person has voted this message useful



s_allard
Triglot
Senior Member
Canada
Joined 3504 days ago

2704 posts - 5424 votes 
Speaks: French*, English, Spanish
Studies: Polish

 
 Message 28 of 136
02 August 2014 at 12:28pm | IP Logged 
emk wrote:
s_allard wrote:
emk wrote:
Or How I Reached 99.8% Comprehension of (Some) French
Fiction While Goofing Off

...

I'm assuming that there is tongue in cheek here. From what I read in the rest of the post, the effort involved in
reaching that level of comprehension of French hardly qualifies as goofing off. That was some serious work!

Well, it was certainly quite a few hours—probably over 400 hours for extensive reading, around 90 hours for Anki
reviews, and some more time spent creating Anki cards, all over the course of about 30 months. But at the same
time, I wouldn't call it "work"—most of that time was spent reading science fiction novels, random French
websites and bandes dessinées, or watching TV series on DVD. I never did any really strenuous vocabulary
work, just easy sentence cards.

...

What I find interesting about the debate here is how we can disagree about the meaning or proper usage of
words in our common language. I won't talk about cheating any more but I'm intrigued by how one can
characterize the time spent studying or "working" on French as goofing off. I checked the verb "goof off" in a few
online dictionaries including Babylon that said:

"(American Slang) be lazy; do nothing (e.g.: "During the school break I am planning to stay home and goof off")

In French, it would be:

"être fainéant; ne rien faire, se prélasser "

For what it's worth, here is the definition from the Urban dictionary:

"To slack off by wasting time idely, sometimes through electronic means like video games, or through alcohol,
doing absolutely nothing, or sleeping excessively. To goof off is to pass the time through the most unproductive
means possible."

Learning a language for nearly all of us is not paid work, I'll admit, but I don't think the time and effort emk spent
learning French or that we spend on our various languages qualifies as goofing off. By the same token, an
amateur musician who spends an hour a day practicing music is not goofing off.

This is not a big issue, of course, but it does illustrate once again how slippery the notion of understanding a
word is. I wonder what most people around here understand when they read of learning a language while goofing
off.
1 person has voted this message useful





emk
Diglot
Moderator
United States
Joined 3606 days ago

2615 posts - 8805 votes 
Speaks: English*, FrenchB2
Studies: Spanish, Ancient Egyptian
Personal Language Map

 
 Message 29 of 136
02 August 2014 at 2:10pm | IP Logged 
Leaving aside discussions of terminology, I'd like to see whether this model makes any useful negative predictions, that is, can it correctly tell us that certain things won't work?

First, here's the diagram again:



And here are two input-based methods that I've seen fail:

Unlikely to Work: Overlearning a few beginner courses without moving on

Some people make great progress with a good beginner course, but afterwards, they never move on to native materials (or to lots of conversations). Instead, they devote a large amount of time to reviewing what they've already studied.

Cheating: With any luck, a good beginner course should provide enough context and explanations so that you can make sense of the text. But you will probably get a very limited sample of all the things the language can do.

Consolidating: I've never seen a beginner course which provided more than a couple hundred pages of content and even that is rare. Even if a student learns this material by heart, there's only so much information which can be extracted from it, and there just aren't enough varied examples to finish the consolidation process.

This approach focuses on the "Cheating" half of the problem, and mostly avoids the "Consolidating" half. I would predict that it leads to slow progress and very long intermediate plateaus. One workaround might be to use multiple courses, including intermediate courses. But even then, it's going to be hard to get thousands of pages of varied native material.

Unlikely to Work: Listening to 1,000 Hours of Incomprehensible Radio News in Mandarin

Every once in a great while, we see somebody try an experiment where they start a new language from scratch, and they try to learn by listening to the radio or watching TV. Now, this might be possible when moving between closely related languages, especially if there were a lot of visual cues and very slow, clear speech. But let's assume that our learner has no special advantages.

Cheating: The learner has skipped this stage entirely, and therefore the audio will remain largely opaque.

Consolidation: With almost no decipherable audio, there's no way to consolidate anything.

I'm embarrassed to say that I actually tried this method with French. When I was trying to improve my French listening comprehension, I left the news radio on a lot. I understood a lot, but often not enough. Here's a note from my log when I was working towards B1:

Quote:
RFI Français Facile was more challenging today, and I didn't understand as much the
first time through. But after the 3rd listen, I got a lot of it. There was a
ridiculously fast report on the soccer championship between Montpelier and
Saint-Germain. The demonstrations in Spain are growing, the demonstrations in Athens
are shrinking, and Germany has a new president, a pastor and human rights advocate from
East Germany. Oh, and Iran has decided to stop selling oil to the UK and France, but
neither country cares, because they don't buy that much Iranian oil, and because the
cold wave has receded. Greece, however, gets 30% of their oil from Iran. In a few
sections of the news report, I could understand plenty of half-sentence fragments but
couldn't tie them together. (It's amazing how some days are so much easier than
others.)

I listened to lots of radio like this. But I only made a modest amount of progress. I didn't really use transcripts, because I hoped to make progress just by listening, and by inferring the rest from context.

But unfortunately, unless you're a news junkie, news radio doesn't always offer enough context. There's a new topic every 60 seconds, the announcers change from program to program, and there are no pictures to sort out ambiguity. So even though I listened a lot, I made only very slow progress. Transcripts would have helped, but using them would have transformed a fun activity into work, so I just kept listening. And I didn't really get anywhere, even with many hours of listening per day.

I didn't see any major improvements in listening comprehension until I discovered DVD box sets. They offered vastly better context: images, a plot, recurring characters, predictable themes, and so on. I watched Buffy (5 seasons), Angel (3 seasons), Ulysse 31 (the whole series), Avatar (3 seasons), and then I began branching out and channel surfing. With the first series, it took me almost three seasons to reach 90%+ comprehension. But with each successive series, it took less and less time to get up to speed, until I could eventually just channel-surf and understand most shows on TV. (And don't forget that I was doing lots of reading, too.)

For me, French news radio didn't offer enough opportunities to "cheat." Parts of the input were semi-automatic, and parts were opaque, but there wasn't really much decipherable input at the border between the two. When I switched to TV, there was actually less automatic input than with radio. But TV offered me far more hints, so I could decipher more.

Personally, I find that the most helpful extensive activities share two features:

1. They're lots of fun, or even downright fascinating. This makes it easy to get enough hours of exposure to consolidate.
2. They offer a relatively large amount of content which is "decipherable" but not yet "automatic," so there's something to consolidate.

Of course, you could discard the "fun" part if you had some other reason to pay close attention to large amounts of input. But I like fun, because (a) it provides a fairly reliable way to judge difficultly, (b) it provides lots of motivation, and (c), well, it's fun.

Edited by emk on 02 August 2014 at 2:51pm

6 persons have voted this message useful



iguanamon
Tetraglot
Senior Member
Virgin Islands
Speaks: Ladino
Joined 3336 days ago

2224 posts - 6707 votes 
Speaks: English*, Spanish, Portuguese, Haitian Creole

 
 Message 30 of 136
02 August 2014 at 2:40pm | IP Logged 
In emk's defense, some learners here present a learning style of something akin to a full frontal assault on a language's form (and not its substance). One need only look at some of the logs to see it- an overwhelming focus on courses and srs- anything to attempt to hammer the language into the brain. What emk seems to be doing with his choice of language- use of the words "cheating" and "goofing off" is to convey a message, at least to my understanding, of "hey, lighten up, you can learn without being so regimented- read a lot and watch a lot of tv and here's how to make it understandable".

In my four years here I've seen some people come on to HTLAL after perusing the usual suspects' youtube videos and think they can learn Spanish in a few months by LR'ing a couple of novels- often with little background in the language beyond ¿Cómo está Ud.?. I've also seen people go nuts over srs, 10,000 sentences, swaedesh wordlists, etc. These types of learners seem to go out of their way to make a difficult task even more arduous than it need be- with few results to show for their efforts. To these types of learners, watching a TV series, reading comic books and novels would indeed be "goofing off". They would see this as not "working" and not "studying"- almost as if doing the types of activities that emk, I, and others advocate is somehow "less worthy" than regimented "study".

Wrapped up in this attitude is also fear. This fear is sometimes based on perfectionism or that native materials are "too hard". Most learners know that they should be engaging the language with native-speakers and native materials. Most want to, but some folks just don't have a clue how to do it. A course holds your hand and guides you to where it wants to take you. In that respect, it's easy. With native-materials and native-speakers there is no guide, one must fill in the blanks relying upon previous knowledge acquired, context based intuition and whatever is available to make the incomprehensible, comprehensible. It's not so easy to do as a ready made, hand-holding course.

What emk has done for beginners is show them a way to do engage with native materials without having to bang their heads against a brick wall only to give up and do Assimil again- and again- "I tried reading X. It was too hard. I had to look up every other word. I'll put it aside until after I've finished X course.". Or, "I tried listening to X. The speaker was too fast. The accent was different than my course. I only caught a few words. I gave up.". This happens all too often and with predictable results.

Who wouldn't want to learn a language "while goofing off"? To many first time language learning beginners this whole mysterious process of learning a language can only be done in a rigid manner- course, anki, more course, more anki, etc. Emk's solid advice has given these learners an annotated road map. Well done, indeed, sir.

There are many members who speak a lot more languages than emk, have a higher level of French than he does and a greater knowledge of linguistics, but they don't have his ease of communication, log popularity or number of useful votes. I've followed his French journey from when he was trying to get his French to B-1. His hard work and effort have been rewarded. I personally believe that he has, indeed, already wandered to C-1. Not bad for a guy who boosted his French while watching dubbed dvd's of "Buffy The Vampire Slayer" and reading comic books. :)
14 persons have voted this message useful



Bao
Diglot
Senior Member
Germany
tinyurl.com/pe4kqe5
Joined 3840 days ago

2256 posts - 4045 votes 
Speaks: German*, English
Studies: French, Spanish, Japanese, Mandarin

 
 Message 31 of 136
02 August 2014 at 2:41pm | IP Logged 
I can see some ways of 'cheating' with radio news even without transcripts:

* you already have a large vocabulary and know how to read, but your listening comprehension is low

* you already know the content of the day's news and can concentrate on how it is reported in the target language

* you make it decipherable yourself (making your own transcript, repeating what you hear phrase by phrase and looking up unknown words, ...) until it becomes transparent enough (this can be really hard and I'd advise people to only do it with things they really want to understand, or if there's nothing else to work with)


I think people who gave me advice IRL only said 'listen to the radio' and not 'you might need to come up with ways to understand it though'.

ETA:
English has this wonderful distinction between hearing and listening to.
If you can't make sense of what is being said in a language I think you should talk about 'hearing it', not about 'listening to it', because I don't think it's possible to sustain your attention on the spoken language (i.e. listen to it) when you can't understand it.

Edited by Bao on 02 August 2014 at 2:45pm

4 persons have voted this message useful



s_allard
Triglot
Senior Member
Canada
Joined 3504 days ago

2704 posts - 5424 votes 
Speaks: French*, English, Spanish
Studies: Polish

 
 Message 32 of 136
02 August 2014 at 4:04pm | IP Logged 
As the one who has most questioned the use of the terminology cheating and goofing off, I want to reiterate my
total agreement with the learning strategy espoused by emk and endorsed by iguamon. It's the terminology that
irritates me but it's not a big deal. And I'm certainly not against fun in learning. Maybe it's my academic
background that prevents me from seeing things in a more popular vein.

When I look at the strategy outlined here, it is in my mind basic common sense. I don't really see another way of
going about acquiring comprehension. Does anybody really believe that you can learn Mandarin or any language
from scratch just by listening to 1000 hours of recordings? Of course it won't work. It never has and never will.
This is akin to trying to learn a language in your sleep. In the same vein, nobody would try to learn to read a
language just by attempting to read a newspaper from scratch.

Similarly, overlearning beginner courses is guaranteed to fail. By definition. If you take a Beginning Mandarin
class in a school or use any of the self-learning products - including, heaven's forbid, Rosetta Stone - and do
nothing else, you will surely fail to attain any kind of proficiency. Everybody knows you have to practice, practice
and practice. As for how to practice, I don't believe that there exists a language teacher that doesn't tell their
students to use native materials of some kind, e.g. DVDs with subtitles, listen to the radio.

If I look at classroom teaching material produced today, there is great emphasis on authentic material. Actually,
this isn't that new. But today's technology takes it to a higher level. As a matter of fact, one American textbook
publisher for Spanish - I can't find the name - offers an integrated tutoring service whereby North American
students can work with tutors in Guatemala using a custom-designed software platform.

The vast majority of self-learners never finish that package they bought with the intention of becoming "fluent"
overnight. But it's not for lack of access to native materials. As a matter of fact, when people ask me about
learning a new language, I always tell them to take a class, preferably a solid immersion class like Middlebury for
those with big bucks. Trying to learn a language on your own is certainly doable, as we all know, but very few
people have the motivation and perseverance for it to work.





1 person has voted this message useful



This discussion contains 136 messages over 17 pages: << Prev 1 2 35 6 7 8 9 10 11 12 13 14 15 16 17  Next >>


Post ReplyPost New Topic Printable version Printable version

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 0.3594 seconds.


DHTML Menu By Milonic JavaScript
Copyright 2019 FX Micheloud - All rights reserved
No part of this website may be copied by any means without my written authorization.