[HN Gopher] An experiment in teaching a language without relying...
___________________________________________________________________
An experiment in teaching a language without relying on translation
Author : toto444
Score : 28 points
Date : 2021-01-24 11:42 UTC (11 hours ago)
(HTM) web link (drdru.github.io)
(TXT) w3m dump (drdru.github.io)
| keiferski wrote:
| Nice idea, but this would be much better as a slightly animated
| video, especially with sound effects. E.g. to emphasize the
| monkey eating a banana, have a sound and animation for it.
| mettamage wrote:
| I don't understand anything and I tried a bit, but IMO this is
| much more effort than normal. A friend of my confirmed you need
| to know hiragana. She has studied it a bit and could follow the
| lessons, whereas I couldn't (knowing next to nothing).
| AnIdiotOnTheNet wrote:
| Well the good news is that the kana are really low hanging
| fruit. You can learn them in a day and as long as you continue
| learning Japanese you will be constantly exposed to them. There
| are only just over a hundred of them and, unlike English, they
| are only ever pronounced one way[0].
|
| Kanji are significantly more effort, with 2k characters to
| learn and multiple pronunciations for each. Interestingly,
| while you will probably hate Knaji at first and wonder why they
| are still used, you will come to appreciate them later as they
| aid reading by breaking things up[1] and disambiguating between
| homonyms.
|
| [0] Ok, sorry, that was the lies-to-children version. Katakana
| are only ever pronounced one way. Hiragana has different
| pronunciation for ha and he when they are used as particles.
| Also there's pitch accent.
|
| [1] Japanese generally doesn't use spaces.
| hombre_fatal wrote:
| I couldn't even tell if the first example was showing how to
| create a conjunction "banana and monkey" or something else like
| "monkey eats banana". Yeah, this would need a lot slower of an
| approach to work for day 0 beginners. Rosetta Stone and
| Duolingo are great examples of how it can be done.
| toto444 wrote:
| The site is definitely not supposed to be the main resource
| used to learn Japanese despite being self contained. It
| exists to complement the learning from a more traditional
| app.
| inetsee wrote:
| This reminds me of "English Through Pictures" by Richards and
| Gibson.
| np_tedious wrote:
| Idk about this. I found myself translating in my head anyway
| axiom92 wrote:
| I worked briefly on a similar idea last year. The problem was to
| collect parallel data for translation without requiring the
| annotators to be bilingual. We achieved it by selecting a set of
| universal pictures (e.g., dog running on grass) and asking the
| native speakers of Hindi and English to caption it. Since the
| concepts in the pictures we selected were common across the
| languages, we were able to collect high quality data.
|
| Paper: https://arxiv.org/abs/2004.11954
|
| Slides: https://madaan.github.io/res/artifacts/pml4dc-practical-
| data...
| gryson wrote:
| But... why?
|
| Translation is not something to be avoided for the adult language
| learner. We should use every advantage at our disposal.
|
| You can spend hours trying to guess the meaning of a word through
| pictures and context (and not getting anywhere far), or you can
| spend a minute reading the translated meaning and then spend
| hours understanding it in context (and reinforcing it, and
| learning its semantics).
|
| I wonder if this idea of "avoiding the mother tongue" stems from
| all the misguided marketing surrounding products like Rosetta
| Stone which advertise "learn like children learn." Spoiler: it
| takes children years and years of near-constant, directed input
| to acquire their native language to a high degree.
|
| Edit: But I can see the value in doing this as a thought
| experiment ("what if translation isn't possible?"), and for that
| it's interesting.
| toto444 wrote:
| > But... why?
|
| I started it for two reasons. First there is close to no easy
| to read content when you begin learning Japanese and there is
| demand for such thing.
|
| The second reason is to see how far I can go like this. It is a
| sort of thought experiment like you say in your comment.
|
| I discovered after I started that Hans Orberg something similar
| for Latin ( https://en.wikipedia.org/wiki/Hans_%C3%98rberg ).
| His Lingua Latina has been quite influential.
| gryson wrote:
| Have you tried more traditional textbooks? I understand
| there's a desire for an easy-to-use interactive app, but I've
| never come across anything remotely as good as a textbook
| like Minna no Nihongo (I used this book 20 years ago when I
| started learning Japanese - maybe I'm biased?).
|
| You can see sample pages here:
| https://omgjapan.com/products/minna-no-nihongo-
| shokyu-1-hons...
|
| It is 100% in Japanese, but it has a companion book with
| translated grammar notes if needed. The majority of the book
| is excellent content in the form of dialogs and readings
| designed for beginners. I suppose the exercises work better
| in a classroom setting...
| hinkley wrote:
| It's not like immersive language classes aren't a thing. Even
| the normal foreign language classes at my university, past the
| third or fourth semester at least, only had 1 day of English
| while handling administravia, and then the entire class from
| then on was in the target language. During summer semester,
| _all_ of the classes were taught this way, even the
| introductory ones.
|
| Part of the difference between children learning something and
| adults learning something is that we have trouble achieving
| Beginner's Mind.
|
| Finding techniques that work fine to be remedial or even
| insulting makes you get into your own way. In the extreme case,
| how do you fill a cup which is already full?
|
| Less philosophically, each language you master gets its own
| language center in your brain, just as each finger you can
| control individually maps to an identifiable region of your
| motor cortex (and if you can't control your pinky, it's because
| they've blended together). Translating back and forth is a
| crutch for most, and a separate skill for others (eg,
| professionals). That's why when you ask your friend, "what did
| those guys say in the background?" you get something vague
| back. Nothing translates directly, idioms doubly or triply so.
|
| What did that guy say? He said he's really mad at that other
| guy, (using a lot of colorful language), and I can say "he's
| mad" or we can go into why calling someone a dog/goat/cow/fish
| in this language is one step down from saying bad things about
| their mother.
| gryson wrote:
| Yes, the 'natural approach' to language learning is
| definitely popular. But in most classroom settings I would be
| surprised if students weren't exposed to the translated
| meanings through homework or the textbook. Class would not
| get very far once students started learning abstract concepts
| otherwise. Of course, translation is just the very first step
| in lexical acquisition - it's still a long journey to acquire
| a word.
|
| >Less philosophically, each language you master gets its own
| language center in your brain, just as each finger you can
| control individually maps to an identifiable region of your
| motor cortex (and if you can't control your pinky, it's
| because they've blended together).
|
| That's not really true, though, and it also depends on what
| aspect of language we're talking about (phonology vs. syntax
| vs. lexical acquisition, for example). In the case of the
| lexicon, there's good experimental evidence that bilinguals,
| when hearing one language, activate lexical competitors from
| both languages (i.e., they are not fully able to inhibit
| activation of one language in listening). The mind is far
| more complex than having different "centers" for different
| things. It's a complex network of signal activation and
| inhibition.
|
| Of course, that doesn't relate to my original point, which
| was simply that language learners should not purposefully
| limit themselves. The incredible amount of time it's going to
| take to learn abstract concepts without translation far, far
| outweighs the possible co-activation of translation
| equivalents.
| rootsudo wrote:
| Kana is really easy to learn.
|
| https://drdru.github.io/stories/1_01_food.html
|
| I found the examples are literally an image abstraction and how
| to read them.
|
| They're what you'd find in a similar Genki Vol 1 book, the emojis
| are cute.
|
| The furigana is invaluable and it helps get the structure in
| place, I do it similar with song lyrics from utanet.
| pchap10k wrote:
| More or less that's right. They're a good step toward seeing
| pictographs as phonetic characters.
|
| When I first starting learning Japanese I took an intensive 7
| week course. We learned Hiragana and Katakana in one week,
| though it took longer to really make that knowledge long term
| available. I forgot a lot of it after completing the course,
| but my first year living in Japan helped restore it all!
|
| Now my kids are learning kana and my oldest kanji. The state of
| the art here is still drilling and repetition. FWIW this
| website is cute, but without learning the characters you have
| to rely on verbal instruction, so as an experiment it only
| works after learning the Kana.
| toto444 wrote:
| Hi OP here
|
| You definitely need to learn Kana beforehand. It used to be
| specified but I decided to remove it a while ago.
| klipt wrote:
| From what I've read, spaced repetition like Anki is state of
| the art for memory intensive subjects like kanji/hanzi.
| zrkrlc wrote:
| I feel like this will only work if you can read hiragana and
| katakana already.
| maweki wrote:
| Not being able to pronounce or hear the words is a problem.
| Associating the words just with the symbols seems very taxing
| to the visual parts of the brain. Also using the
| voice/vocal/speaking parts of the brain seems advantageous.
| tsimionescu wrote:
| Not to mention, even if you could manage to remember the
| symbols that way, it would still be next to useless, since
| you still couldn't SPEAK Japanese at all.
|
| The site creator has stated elsewhere though that the site is
| intended for people who already know hiragana/katakana (the
| phonetic parts of the Japanese writing system). If you do
| know those, it could help as practice for kanji and phrases.
| toto444 wrote:
| Hi
|
| I am the OP. I just discovered that my submission has made it to
| the front page.
|
| A quick word on the rationale for not using translation.
|
| All of the classic apps (Anki, Duolingo, Memrise, ...) are to
| some extent translation based. Translating has three scalability
| issues.
|
| First, if you want to be inclusive and your product to be used by
| everyone you have to translate the whole product in a long list
| of languages. I imagine there is a ton of people who do not speak
| English and would love to learn Japanese !
|
| I don't really know how to express the second limitation better
| than by saying 'translation is not additive' or it does not
| 'scale up'. What I mean by that is that if you translate at the
| word level (such as with Memrise), you learn words but somehow
| you do no learn the language (to be fair is the point of the app
| and that works fairly well). However a language expresses itself
| through sentences and translating a sentence is not the same as
| translating each one of its word. So you are sort of stuck at the
| word level and can't scale up to the sentence level by combining
| the low level components that words are.
|
| Third, translating sentences does not scale well due to the
| complexity of natural languages. If like Duolingo you base your
| app on translating sentences you face all the poblem faced by
| Duolingo.
|
| How do you translate Yuan habananawoShi beru ? There are a lot of
| potential translations. Here are some I can think of right now :
|
| The monkey eats the banana.
|
| The monkeys eat the banana
|
| The monkey will eat the bananas.
|
| The monkeys are going to eat the bananas.
| 616c wrote:
| This is basically the Uplan method, no? The Israelis pioneered
| this long ago, or so I have been told, by relatives who were
| taught Hebrew in the 70s when studying there. Hebrew as spoken
| today (aka Modern Hebrew) was a new language that won by vote
| (to IIRC Russian, Yiddish, German and others) when the country
| had to decide on a national language, and they needed a way to
| teach people uniformly.
|
| https://www.ulpanor.com/
|
| That being said, as they learned via Ulpan as adults in their
| 20's and spent only 1-2 years there, none of them speak beyond
| a few words of Hebrew once I was in their lives decades later
| up until now. Twas quickly forgotten. I went through
| conventional language education in Arabic from early teens (it
| was my desired hobby because of my obviously vibrant social
| life, hehe) to college and speak it so-so until this day, but I
| have many friends and family of my own who speak Arabic.
|
| So more than anything: increased exposure is key and few people
| will correct humor you with your method if it takes longer as
| an adult, and it definitely will! haha
| tsimionescu wrote:
| I think that one problem that eventually happens with this
| approach, which is a problem I had with Duolingo as well, is
| that very soon after learning a few words you need to start
| learning basic grammar. In these approaches that don't explain
| the grammar at all, you need to spend effort pattern matching
| to pick up the right grammatical rules to construct your own
| sentences, and you are never sure that you're doing the right
| thing.
|
| For example, in Yuan habananawoShi beru, how should I
| understand what ha and wo _mean_? When should I use them? I can
| more or less easily understand that Yuan means monkey, banana
| means banana, and Shi beru is related to eating. But to
| understand that ha indicates the topic and wo indicates that
| the word before is having an action done to them, I would have
| to see a LOT of examples.
|
| I think a little grammatical explanation goes a very long way
| in terms of the number of examples you have to show.
___________________________________________________________________
(page generated 2021-01-24 23:00 UTC)