|
 |
The Cheating & Consolidating Method | ||
Share with:
| ||
 Language Learning Forum : Learning Techniques, Methods & Strategies |  | |
| Expugnator Hexaglot Senior Member Brazil Joined 5165 days ago 3335 posts - 4349 votes  Speaks: Portuguese*, Norwegian, French, English, Italian, Papiamento Studies: Mandarin, Georgian, Russian | Message 89 of 136 05 August 2014 at 9:38pm | IP Logged | ||||||||
|
The past discussions made me think a lot. I had some insights about what I had been doing and how to use emk's theory to explain the issues I've been having. I couldn't reply at the moment so I missed the chance to answer, but what I can recall is as follows:
@Serpent: GLOSS is very good but I could only use it at home, and that's not where I do most of my study o.O So, I came to the conclusion that with Russian, even though it has an abundance of resources, it is hard to find n+1 material in a consistent way. The textbooks themselves don't respect this principle. Assimil introduces too much vocabulary all at once. Since Russian has so much alien vocabulary, Assimil attemps to cram in the same amount of vocabulary for Russian one would acquire for French or Spanish, but forgets that this vocabulary is almost entirely new, unlike in Spanish or French where 50% would be cognates (75% for those coming from another Romance language). So, this attempt to make up for the vocabulary gap turns out to be counterproductive,even more so that the words don't get repeated that often as in the 'good' Assimils. This issue doesn't happen in Assimil Chinois: the amount of words is kept at an optimal ratio. We may say Assimil"respects" Chinese difficulty but doesn't respect Russian dfficulty, because, after all, when we're talking about brand different roots to memorize, long, verbose, not-so-phonemic cyrillic Russian words may require as much effort as bisyllabic tonal ones with hanzi. This particular issue with Assimil Russian concurred to me not 'finishing' consistenly the A2 level. Then let's consider intermediate texbooks. They're a huge gap! from beginner ones, even considering one has fully understood the beginner ones. I just checked Assimil Perfectionnement Russe and at my current stage i'm indiferent between using Perf. Russe or a novel with a translation from an unknown story, that is, I find both utterly difficult. A parallel reading from a known story would be slightly easier, but even at this point the huge amount of unknown words in Russian would require me to go too slowly through each paragraph. Even Ilya Frank's method is still a bit hard, that is, I may only understand 30% of the vocabulary. Bear in mind, that is totally different from coming from a basic Assimil in a Romance language, where you probably go from an initial 75% to a 90% comprehension of written texts with much less effort. For the moment, I decided I'd work on other intermediate textbooks, such as Colloquial Russian 2, before tackling Assimil. I'm already and will keep working on native Russian materials, such as the Agatha Christie's novel, and I expect synergy to make me enjoy this reading more by getting to know more and more words. I will try to cheat more to make text at CR2 decipherable, and this includes OCRing text and finding the translation the course didn't provide me. The fact the text didn't provide it only made things more difficult to me and actually damaged my learning, because having to look up individual words and then losing track when you get back to the sentence totally prevents me from learning vocabulary in a proper and contextual way. Comparing my learning from a Russianpod lesson, where evertything is translated, even the sentences used in grammar explanations, and CR2 where I get texts untranslated and have to figure out things on my own suffices to notice how much better I retain stuff from the Russianpod lesson, which is actually longer. This brings up another issue when I realized how a textbook harms instead of helping out my studies when it keeps doing what had been working properly. I'm talking of Basic Course in Estonian now. The initial conversations all have translations, and new words and grammatical features are explained with literal translation (the FSI style most people is familiar with). Then, probably due to lack of space, they don't provide translation for the exercises and the final conversations. And that only turns half the content of the lesson in a nuisance! Instead of studying the whole lesson in a practical, consistent, encouraging and effective way, I have to struggle through half the lesson, checking words manually and losing the meaning of the sentence and the context when I come back. So, today I decided I'd OCR these pages with Estonian-only sentences, and as a result, I noticed I could learn much better, even aspects I was in doubt about, when pasting those OCRed texts in Google Translate. Therefore, 'cheating' with technology helped making my learning experience much more efficient. BTW, today's lesson was only reviewing, 8 pages of sentences, no translation, no audio. It allowed me to notice clearly how 'not cheating' can be much more tiresome and, most important, much less efficient to me. Edited by Expugnator on 05 August 2014 at 9:42pm 2 persons have voted this message useful | |||||||||
| s_allard Triglot Senior Member Canada Joined 5429 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish | Message 90 of 136 06 August 2014 at 2:08pm | IP Logged | ||||||||
This sort of reaction proves exactly why "cheating" is, in my opinion, a poor choice of terminology. As is to be expected, some people get distracted by the negative connotations and no longer pay attention to the real message, thereby wasting their time and everybody else's. Now on to some serious stuff. When I survey the discussion so far, I see lots of great ideas and suggestions for what I would call tips and tricks of more efficient language learning. My own conclusion is basically explore and experiment to see what works for you. If something isn't working or you find it boring, then drop it and try something else. Indeed, as a direct result of this thread, I've seriously restarted working with Anki for my Spanish. I've created a couple of specialized decks focusing on areas of Spanish grammar that are difficult for me. For example, I'm now populating a deck on the imperfect subjunctive and another one on complex constructions. What I would draw attention to is the importance the social context of learning and actual meaningful contact with the language. Immersion where you are constantly stimulated by comprehensible (and incomprehensible) input and where you can immediately use the language is very important. And having a significant relationship with a speaker of the language is also very important. A classic situation that I see constantly here in Montreal, Quebec, is the English-speaker who marries a French- speaking Québécois and is raising a family in French. Quick success in learning French is 100% guaranteed. Obviously, most of us are not in that kind of situation, but I think a deliberate strategy of personal contact with the language through short trips to the country where feasible and interacting with immigrant communities should be considered. 1 person has voted this message useful | |||||||||
 emk Diglot Moderator United States Joined 5531 days ago 2615 posts - 8806 votes Speaks: English*, FrenchB2 Studies: Spanish, Ancient Egyptian Personal Language Map | Message 91 of 136 06 August 2014 at 3:54pm | IP Logged | ||||||||
Yeah, I sympathize completely. Finding decipherable "i+1 input" is is much harder for Egyptian than French. I have a couple of strategies for dealing with this: 1. I spend a fair bit of time (and sometimes money) trying to find the right kind of native materials. This usually involves finding more materials than I need, and then digging through them until something "clicks." For Egyptian, I'm using Assimil, the St. Andrews corpus (free!) and a dubious bilingual edition of a classic tale. But I also tried other things which didn't work out, including a beautiful copy of The Book of Going Forth by Day, bilingual editions of Budge from Dover, etc. 2. There's more than one way to decipher a text and turn it into "i+1 input." Some texts can be deciphered just by reading them (typically around 95+% initial comprehension). Others can be deciphered by guessing and skipping (around 70% initial comprehension). But it's possible to tackle much harder texts using an interlinear translation, a concise grammar summary (like Iversen's green sheets), and occasional dictionary lookups. Ultimately, it doesn't seem to matter how I decipher stuff; it all consolidates the same way. 3. When decipherment is really difficult, it helps to re-read what I've deciphered multiple times. This doesn't help as much with consolidation as new material would, but sometimes all the other choices are worse. This is especially true in the beginning, and it's why I go through each Assimil lesson 8–12 times, and it's why I use Anki. There's some other stuff I haven't mentioned: It's no good just understanding the L1 version of an interlinear text. Instead, I need to use the English version to "understand" the TL version directly, as best I can. If decipherment was really hard, it's good to cover up the L1 version and read the TL version directly a couple of times.
I love this! It's sneaky, it's underhanded, it ignores the "How to Use This Course" directions—and it produces decipherable text that helps you consolidate your understanding of the language. I'm really glad to hear that this is working out for you so far. 3 persons have voted this message useful | |||||||||
| emk Diglot Moderator United States Joined 5531 days ago 2615 posts - 8806 votes Speaks: English*, FrenchB2 Studies: Spanish, Ancient Egyptian Personal Language Map | Message 92 of 136 06 August 2014 at 11:56pm | IP Logged | ||||||||
|
OK, since various folks have asked about measuring comprehension, I decided it would be fun to try a little experiment. I took a short passage from L'Homme qui plantait des arbres, a short, public domain story in French (so we don't get the admin in any trouble). This is actually a pretty nice beginner story, especially if you skip a few hard words.
I chose the first 347 words, which is roughly one paperback page. As usual, the first page contains an unusually high number of rare words—this happens because the author is setting the scene and writing more descriptive prose than will typically appear later on. I've provided a translation of each sentence in English, and I've marked decipherable text in blue and opaque text in red. I've also added a note about each colored word. Everything is pretty much automatic at this point. 94.5% automatic, or at least transparent (328 words) 3.5% decipherable (12 words) 2% opaque (7 words) ??? incorrect Normally I would do slightly better than this, because there would be less geography and botany in a typical novel, at least after the first few pages. I rarely add botanical terminology to Anki, because I don't actually care much about it unless I need it in conversation with my family. Also, when I'm translating for public consumption, I'm more careful than I'd be if I were reading for pleasure, and I'm less likely to miss a genuinely opaque word.
When I read my first actual book in French, there would have been massive amounts of blue and red all over this page. Anyway, if anybody wants to point out my mistranslations, I'll be happy to update my percentages! UPDATE: Thank you to s_allard for catching my mistranslation of clocher "bell tower" as cloche "bell." Please chalk one more word up to "opaque." Edited by emk on 07 August 2014 at 11:04pm 4 persons have voted this message useful | |||||||||
| tarvos Super Polyglot Winner TAC 2012 Senior Member China likeapolyglot.wordpr Joined 4706 days ago 5310 posts - 9399 votes Speaks: Dutch*, English, Swedish, French, Russian, German, Italian, Norwegian, Mandarin, Romanian, Afrikaans Studies: Greek, Modern Hebrew, Spanish, Portuguese, Czech, Korean, Esperanto, Finnish | Message 93 of 136 07 August 2014 at 11:01pm | IP Logged | ||||||||
|
One comment on the lavender - here the reference isn't just a descriptive element.
Lavender is actually the symbol of the Provence region in France. Given the setting, mentioning lavender in the description is pretty much inevitable. It's also a cultural reference. It would be like going to Mexico and not mentioning the cactuses. Edited by tarvos on 07 August 2014 at 11:07pm 3 persons have voted this message useful | |||||||||
| s_allard Triglot Senior Member Canada Joined 5429 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish | Message 94 of 136 07 August 2014 at 11:09pm | IP Logged | ||||||||
Since I've started a separate thread on this issue of comprehension, I'll be very brief. Although it is not stated as such, I'm going to assume that the level of comprehension here is 93.4% that corresponds to the number of words that were "transparent" or "automatically" understood out of 347. Let me first point out that emk has undoubtedly an excellent reading command of French and has understood the text very well. The only way we can verify this understanding is through the translation. This where there are a number of issues. I'll just make two remarks about the excerpt above: 1. dépouillé can mean skinned in certain contexts but not here. It would be more like "without", "lacking of" or "devoid of" 2, égoïsme here is not egotism. This is a bit of a false friend here. The sense here is more "selfishness" which is the most common translation. There are minor issues of translation as well but they really don't change much in terms of meaning. In most of the passages, I have issues with what I would call the understanding of this text. In this particular excerpt two important content words seem to be misunderstood. But the point of all this is not to say that the true figure should be lower, e.g. 85% instead of 94.5%. The significance of this is that one can never tell when a person has understood or misunderstood a text. In this reader's mind all these words were automatically understood. It's only when we look at the translation that certain doubts may arise in the mind of an observer. 1 person has voted this message useful | |||||||||
| kanewai Triglot Senior Member United States justpaste.it/kanewai Joined 4888 days ago 1386 posts - 3054 votes Speaks: English*, French, Marshallese Studies: Italian, Spanish | Message 95 of 136 07 August 2014 at 11:25pm | IP Logged | ||||||||
|
I'm amused that you missed lavandes but had no problem with comme un vieux nid
de guêpes! I wish I could compare, but I've read the text recently, so it's hard to judge after the fact which words I had trouble with. I like the idea of translating without a dictionary & then checking the results; it would make an interesting mini-challenge (paging Ms. Solfrid ... ) 1 person has voted this message useful | |||||||||
| emk Diglot Moderator United States Joined 5531 days ago 2615 posts - 8806 votes Speaks: English*, FrenchB2 Studies: Spanish, Ancient Egyptian Personal Language Map | Message 96 of 136 08 August 2014 at 12:25am | IP Logged | ||||||||
I would certainly not consider this to a "finished" translation. I deliberately made the base translation rapidly and without a dictionary, to simulate ordinary reading as best I could, and I marked words that were potentially tricky for further review. Also, this translation is deliberately over-literal, and it favors the most central definitions of a word over those definitions that a dictionary would mark as "by analogy." For example, I perfectly well aware that the phrase "If this action is skinned of all egotism" is absolutely dreadful English, and I would never include it in a finished translation. :-) The word "skinned", for example, has a much stronger connotation than dépouillé. In a finished translation, a better choice would probably be "stripped", which is a more neutral and idiomatic choice in English. Your choice of "devoid of", however, is probably the ideal solution. This is a tricky issue when translating: It's not enough to know what the word means; it's also necessary to know, for example, which metaphors are "dead," and no longer carry any metaphorical implication for native speakers. Translating a dead metaphor literally in a finished translation is almost always a mistake, because it makes a plain style overly exciting. But for this exercise, I went for a deliberately over-literal translation, because a finished translation would make it too easy to hide my errors. A translation which captures both literal meaning and style is very hard to pull off, and I certainly can't do it in a single quick pass. (And as for égoïsme, well, WordReference gives me "egotism, selfishness", which is pretty much what I would have personally guessed. Honestly, I consider these words to be near-synonyms in English. Or if you don't trust WordReference, try a corpus search.) But now I'd like to bring this back to my larger point: This is an ongoing process. When I read my first hundred pages of real French, I was guessing like crazy. Now, I'm worrying about how literally I should translate dead metaphors. After another 10 million words or so, with a bit of strategic use of SRS and pop-up dictionaries, and my French will be even better. But here I run into a problem: Even relatively good native students may take a decade to read 10 million words, with much help from libraries and bookstores. As usual, Itchy Feet summarizes the problem better than anyone. As a language learner, I embrace and accept imperfection. I wear my translation errors with pride! (And thank you once again for clocher.)
Remember, I use my French in real life! I'm perfectly happy to lump 99% of flowers together under fleur, but a nid de guêpes can be a very lively subject of immediate interest. :-) Edited by emk on 08 August 2014 at 12:28am 1 person has voted this message useful |
|  | 
Printable version
|
| Copyright 2024 FX Micheloud - All rights reserved No part of this website may be copied by any means without my written authorization. |
