My poor little brain has been struggling with something for awhile now. Funnily enough I'm not talking about Japanese but rather English and using it to learn Japanese in particular.

It seems there are two styles when learning Japanese; quick and dirty (Ie using English) and purism (ie everything has to be as true to it's natural source as possible)

it's just that sometimes I see purism get a bit out of hand. It has it's place but for me and this is only personal experience and opinion, it's fine to perfect your knowledge but realistically it makes sense to do that last. Kinda boggles my mind how people at times are just trying to "go monolingual" because either khatz said it, it's cool, it's helpful or some crazy mix of all 3.

I've also seen it applied to learning kanji. People going out of their way (not really talking bout this community here) to stay as pure as possible to the real meaning of radicals when making mnemonics. Nothing wrong with it per se but I feel sometimes I see it taking a lot longer than a quicker "dirtier" method.

How realistic is using a monodic? I do wherever possible, yes it's helpful, it's also a bit slow sometimes and extremely slow when too much has to be looked up. Sometimes I think people are a tad naive, I know I was, when it comes to expecting at a certain level it will just become magically possibly.

Upon thinking about it, whatever word you look up will reference a related subset of vocabulary. This is advantageous and also where the problem lies. Until your vocab is anything other than gigantic there will nearly always be lookups within the definition because every definition contains different words that span the entire language.

It is a wicked method of learning collactions for words and I'm definitely not bashing it. Maybe just calling for a little bit of a reality check on the ajatt dream.

I really don't think English will harm your Japanese as much as people seem to think. It's kinda treated like the plague but it's a cold at best, you get over it pretty quick. I find as I'm able to begin comprehending longer sentences in realtime that I don't need to translate the to English to understand them. Just like heisigs keywords it's falling away. In the meantime it's a fast way to learn new things that real exposure is doing a wonderful job filling in the details.

English - help or hindrance?

I agree with what you're saying. I don't think English hurts your Japanese, and I still use it for definitions and what not. Once I can remember what a word means, or even before that, I don't use the English definition anymore. I definitely don't translate Japanese to English unless it's just for the fun of translation.

Monolingual dictionaries are okay sometimes, but other times they kinda give me a headache. I just want to know what a word means ASAP so I can get on with what I'm doing. I don't want to figure out which section actually defines the word, which part is for sentence examples, which is just telling me what part of speech the word is, or which number has the relevant definition...

Just recently I've been using Tanuki to make some vocab lists to re-enforce words that have become a bit shaky and to learn new words. I'm using the Tanuki definitions which are very short and simple and rarely use words I don't know.

I use English for probably 95% of my definitions. It's not a problem because it's just there as a quick reminder/a tool for checking that you were correct. You're not USING the English when you're reviewing, most of the time you won't even look at it.

I think it's good to go monodic, but you have to be really good to get a better grasp of the words using that than using English. If you're learning the word 黒い, "black" as a definition does just as good as a monodic definition. In fact, it probably does it better.

I was thinking about this the other day, actually. Sometimes the definition in a J-J dictionary is really helpful for pulling together a concept that in English is spread across multiple words and giving you a clearer handle on things. On the other hand for a word like 鉱物 I have to say that I suspect that using a definition of "天然に産する無機物。ほぼ一定の化学組成と通常ほぼ一定の結晶構造を持つ固体。まれに非晶質のものや液体（水銀）もある" leaves me with a much less clear grasp of the word than just saying "mineral".

At the moment I'm experimenting with:
* English single-word definition when there is a clear near 1-1 match
* Japanese sentence-length definition when it's straightforward
* English again when things get more complex

I agree, using English definitions is probably not going to harm your Japanese. I almost never used a monolingual dictionary when I was learning French. Still, I think using a monolingual dictionary is a good habit to get into, mainly because you're doing more thinking in your target language. I think pm215's method sounds pretty good.

What I'm doing now is usually using Rikaichan as I go through a transcript or subtitles file. When there are new words, I'll usually look them up on Sanseido and/or Yahoo!'s monolingual dictionaries. I then usually read those definitions using Rikaichan. This often clarifies the original English definition. Then I put the English definition into the SRS. I'm hoping that this will help me start to use more monolingual definitions sooner.

To the OP: have you read Kató Lomb's book Polyglot: How I Learn Languages? The whole thing's available for free online. I recommend reading the chapter called "Dictionaries: Crutches or Helpful Tools?" I think she makes a strong case for using monolingual dictionaries, and it obviously worked really well for her. She basically says that the extra effort involved in using a monolingual dictionary means that you will probably remember the word better than if you had looked it up in a bilingual dictionary.

One thing I've actually been thinking more about recently is how including translations in sentence cards could be detrimental. I'm not really sure if it is or not, but in a way I think that having part of your answer/expectation when reviewing to be to translate to English could be detrimental in that it encourages translating in your head. A lot of the time I find English translations really annoying and not very good at capturing the original Japanese. What I've started to do is stop including English translations when I can understand a sentence well without them, like when a card has no new grammar in it. I'm only mining drama at the moment and watch everything in Japanese with English subtitles first, so I've already seen a translation of the sentence, so I feel like it's not important to include it in my cards. I'm finding that I prefer doing it this way, since including English translations had been really annoying me.

I think it's harmful if you rely on it. I use English as a prompt for production vocabulary cards. I still include the English on the recognition one for association, but I include a monolingual definition and example sentences as well. I can't count the number of words where the English covers less than 50% of the meaning of the word which I get from the monolingual definition. It's no wonder my vocabulary was crap before I decided add to the English.

You either need to read the monolingual definition, or see a lot of really good example sentences. You're definitely not going to be able to use complicated words from electronic dictionary style English definitions alone. It gives you nothing more than the gist.

pm215 Wrote:I was thinking about this the other day, actually. Sometimes the definition in a J-J dictionary is really helpful for pulling together a concept that in English is spread across multiple words and giving you a clearer handle on things. On the other hand for a word like 鉱物 I have to say that I suspect that using a definition of "天然に産する無機物。ほぼ一定の化学組成と通常ほぼ一定の結晶構造を持つ固体。まれに非晶質のものや液体（水銀）もある" leaves me with a much less clear grasp of the word than just saying "mineral".

At the moment I'm experimenting with:
* English single-word definition when there is a clear near 1-1 match
* Japanese sentence-length definition when it's straightforward
* English again when things get more complex

That's interesting. I would have thought the English single-word definitions would be the worst since that makes it really easy to translate inside your head. If you stuck to long english definitions only, I think you'd end up working a bit harder, but wouldn't end up translating internally.

speaking of rikaichan and original look-ups... that is precisely why i study anki online on firefox instead of the stand-alone program... anytime i don't get something i just run it over with rikaichan until i do...

I've always been of the 'treat English definitions like self-dissolving stitches' persuasion--i.e. to make the transition organically because I really don't think there's a 'virtuous circle' amplification from forcing monolingual as Khatzumoto and others claim, but I've definitely taken note of the 'uncanny valley' effect over time where the Japanese is foreign yet familiar and the English is familiar yet foreign, and even though I feel it receding as I progress, I don't think it'd hurt to give it a push now and then. I've got a few ideas there but think I'll wait a while before suggesting, plus many others who've been interested in how to go monolingual have come up with different ideas that could be applied less radically to giving oneself a nudge.

wccrawford Wrote:That's interesting. I would have thought the English single-word definitions would be the worst since that makes it really easy to translate inside your head. If you stuck to long english definitions only, I think you'd end up working a bit harder, but wouldn't end up translating internally.

To a first approximation I don't translate internally, except for the cases where I've only half remembered a word and am mentally fishing for the meaning. Once you've encountered 鉱物 a few times you don't stumble over it any more, you just think こうぶつ and the English doesn't particularly intrude itself. You'll notice that I said "where there's a clear 1-1 match", ie where having the English word in your mind doesn't cause you to get a wrong idea of the meaning. I wouldn't try to do that for something like やっぱり, which is better learned from lots of sentences involving it.

I'm sure I learnt 大学 from some word list which said "university" once, many years ago; that doesn't mean I think "university" every time I read it these days. I don't think "educational institution usually attended by 19-21 year olds" either, or a definition in Japanese. When you actually know a word you don't think about it, you're thinking about the whole sentence of which it's just a part. </captain-obvious>

I usually don't use English or German definitions (except for maybe the first few hundred cards) as Japanese definitions work pretty good for me in maybe 98% of all cases. The rest, e.g. names of animals, plants etc. I usually learn by putting a picture on the question side and the Japanese word on the answer side. Works just fine.
That being said, a number of times the Japanese definition manages to include details that could not be conveyed by any translation. Translations might not be harmful per se, but I kind of don't want to miss out on how some concepts are conceived differently in Japan.

Something which is essential to begin with cannot possibly be considered a hindrance.

Using a monolingual dictionary is clearly better if you can understand the definitions, but this idea that J-E dictionaries are the antichrist is really quite ridiculous.

I love the idea of a j-j thesaurus. I have links to two but it's a bit of a hassle having to always load a webpage. If I could get one to work with stardict that'd be brilliant.

mezbup Wrote:I love the idea of a j-j thesaurus. I have links to two but it's a bit of a hassle having to always load a webpage.

Do you use Firefox? If so, right click in the search box, then click "Add a Keyword for this Search". Then you can search the site(s) using your address bar by just typing the keyword before what you want to search for, e.g. if your keyword is "th" and you want to search for "食べる", all you need to type is "th 食べる".

harhol Wrote:

mezbup Wrote:I love the idea of a j-j thesaurus. I have links to two but it's a bit of a hassle having to always load a webpage.

Do you use Firefox? If so, right click in the search box, then click "Add a Keyword for this Search". Then you can search the site(s) using your address bar by just typing the keyword before what you want to search for, e.g. if your keyword is "th" and you want to search for "食べる", all you need to type is "th 食べる".

This works even better in google chrome. If you've been to the site and used it's search engine, google chrome has already saved it. Write the start of the website in the adressbar, press tab and enter the search word.

You can also manually tell google chrome to allow a keyword to be used as well.

Say you've been to en.wikipedia.org and searched something so google has it saved. Writing en.wi, pressing tab and writing Japanese will search wikipedia for "Japanese". If you've set en.wikipedia.orgs search engine to the w keyword in options, you can write w, press space and write "Japanese" with the same result.

in my view, you cannot learn a foreign language IN that foreign language, at the very outset. babies can, but once you grow out of "babyhood", you lose the Language Acquisition Device, which allows you to just soak up language from the language itself. You can listen to a tape a million times, and may learn to distinguish words, and even guess some meanings, but a non-baby will not be able to miraculously learn the language. I'm not well informed on this, but it puzzles me that many people who go to Japan to do english teaching that I've talked to can't actually speak Japanese. It makes me wonder how they actually expect the poor kids to learn! it would be like dumping an Australian primary school student in a classroom in China and crossing your fingers.

so use English at the start, definately. Use it a lot, in fact, if you learn better that way. I'm intermediate level in french, and I still use English.

vinniram Wrote:in my view, you cannot learn a foreign language IN that foreign language, at the very outset. babies can, but once you grow out of "babyhood", you lose the Language Acquisition Device, which allows you to just soak up language from the language itself. You can listen to a tape a million times, and may learn to distinguish words, and even guess some meanings, but a non-baby will not be able to miraculously learn the language. I'm not well informed on this, but it puzzles me that many people who go to Japan to do english teaching that I've talked to can't actually speak Japanese. It makes me wonder how they actually expect the poor kids to learn!

I always assumed they functioned as teaching assistants providing pronunciation and conversation practice, but maybe I'm wrong. When I was around 13 in school I took a couple Japanese classes but the teacher's lack of English was frustrating for me when I wanted to ask for basic explanations to things.

vinniram Wrote:I'm not well informed on this, but it puzzles me that many people who go to Japan to do english teaching that I've talked to can't actually speak Japanese. It makes me wonder how they actually expect the poor kids to learn!

They're ALTs, or Assistant Language Teachers. It varies from job to job, but they're basically there to smile & nod. Great work if you can get it. That's why a billion people apply for JET each year. I planned to apply, but my old university had a server crash and I couldn't get hold of an academic transcript in time (partly my fault for waiting until two weeks before the deadline >_>). If you want to be an actual teacher, I think you need at least 5+ years previous professional teaching experience in addition to recognised language qualifications (i.e. JLPT).

mezbup Wrote:I love the idea of a j-j thesaurus. I have links to two but it's a bit of a hassle having to always load a webpage. If I could get one to work with stardict that'd be brilliant.

I'd be wary of monolingual thesauri. I think they could maybe help with understanding monolingual definitions, but using them by themselves could easily lead to misunderstandings. Gethin and Gunnemark write about this in The Art and Science of Learning Languages (1996):

Gethin and Gunnemark Wrote:The only thing worse than a monolingual dictionary is a thesaurus. The native speaker sifts the 'synonyms' she finds in a thesaurus, and discards most - or even all - of them. She is able to do this precisely because she already knows exactly what they mean and can accept or reject accordingly. If she is not sure of the meaning and use of a word, she does not dream of using it. A foreign student cannot possibly discriminate in this way.

Source: http://www.lingua.org.uk/voc.html#v58

(I think the whole section on dictionaries in that article is actually very relevant to this thread/debate and that they make some good points about how monolingual dictionaries can sometimes be less helpful than bilingual dictionaries. I don't think their arguments actually prove that using monolingual dictionaries is inherently bad/detrimental, though. In the end, I think a lot of what they're arguing about comes down to the quality of the dictionary. I don't think it's surprising that a patronizing learner's dictionary that oversimplifies definitions could be less useful than a more detailed and accurate bilingual dictionary. So, some of their arguments don't really have anything to do with whether the dictionary is monolingual or bilingual, IMHO.)

No. 鉱物 and "mineral" aren't the same. They're very very veeeeeeery different. The former is only found in rocks, earth etc. while the latter can be in food and drink. They might be referring to the same stuff in a scientific sense, but linguistically they're totally different.

By the same token, 大学 and "university" are different in many ways. りんご and "apple" are also different. Just because they mean the same thing doesn't mean what pops up in native speakers' minds is also the same.

I don't know if using translation helps learn a word. Maybe it does. Maybe it's detrimental. But I don't think it's important. I think you know a word only if you can come up with typical situations where the word is used and natural sentences native speakers would say in such situations. If you can't, you don't know the word yet. Whether it's translation or a monolingual definition doesn't matter. Ask yourself if you can give natural, typical example dialogues. You know the word only when you can confidently say, "Yes. I sure can."

In this sense, the difference between knowing an equivalent word or a monolingual definition is insignificant and irrelevant. They have absolutely nothing to do with "knowing a word."

Also, I do believe that the critical period thing is plain nonsense. It's an excuse for whiners. Maybe there are certain scientific differences in our language learning process between kids and adults. But it doesn't matter at all. I have more than one friend who started learning foreign languages after puberty and achieved fluency to the extent that no one can distinguish them from native speakers. What's actually happening in our brains might be different, but we can learn language the same way. The possible scientific difference means nothing to learners.

IceCream Wrote:As soon as it's possible, you should move to J-J only definitions in your srs wherever possible, or, for certain types of words, J thesaurus entries are better.

I try to use J-J as much as I can and I agree that the thesaurus entries are very useful. It's a lot easier than trying to read through some of the definitions - especially when there are a lot of different meanings. You also get the benefit of seeing more new words and trying to guess their meaning.

But sometimes, if there are too many words that I don't understand I will use J-E and then go back to the J-J to see if I can understand it better.

I disagree with the idea that using English when learning Japanese is bad. It depends on your preference and your level. If it takes you forever to look up the meaning of a word because you have to look up the meaning of more words and so on then I think its counterproductive. But I think that using the J-J dictionary as much as possible is the way to go.

magamo Wrote:No. 鉱物 and "mineral" aren't the same. They're very very veeeeeeery different. The former is only found in rocks, earth etc. while the latter can be in food and drink. They might be referring to the same stuff in a scientific sense, but linguistically they're totally different.

By the same token, 大学 and "university" are different in many ways. りんご and "apple" are also different. Just because they mean the same thing doesn't mean what pops up in native speakers' minds is also the same.

I don't know if using translation helps learn a word. Maybe it does. Maybe it's detrimental. But I don't think it's important. I think you know a word only if you can come up with typical situations where the word is used and natural sentences native speakers would say in such situations. If you can't, you don't know the word yet. Whether it's translation or a monolingual definition doesn't matter. Ask yourself if you can give natural, typical example dialogues. You know the word only when you can confidently say, "Yes. I sure can."

In this sense, the difference between knowing an equivalent word or a monolingual definition is insignificant and irrelevant. They have absolutely nothing to do with "knowing a word."

Also, I do believe that the critical period thing is plain nonsense. It's an excuse for whiners. Maybe there are certain scientific differences in our language learning process between kids and adults. But it doesn't matter at all. I have more than one friend who started learning foreign languages after puberty and achieved fluency to the extent that no one can distinguish them from native speakers. What's actually happening in our brains might be different, but we can learn language the same way. The possible scientific difference means nothing to learners.

Ha, I was just going to make fun of the Critical Period Hypothesis and the LAD/UG theories and post links to other cognitive science views on the topic. Guess I'll save my energy. I've been reading some interesting stuff in MacWhinney's unified model about resonance that made me want to read more and devise ways to incorporate it into timeboxing (well, my particular strategies for timeboxing esp. as they relate to overall SRSing strats), but I'll save it for a rainy day...

magamo Wrote:No. 鉱物 and "mineral" aren't the same. They're very very veeeeeeery different. The former is only found in rocks, earth etc. while the latter can be in food and drink. They might be referring to the same stuff in a scientific sense, but linguistically they're totally different.

By the same token, 大学 and "university" are different in many ways. りんご and "apple" are also different. Just because they mean the same thing doesn't mean what pops up in native speakers' minds is also the same.

Well, when I think "mineral" I think "the thing you find in rocks and earth". "mineral water" and "vitamins and minerals" are kind of special cases. So the concepts are basically close enough even if the match is a bit fuzzy round the edges. For a first cut at understanding a word that's really the best you can hope for -- the only way to fix the fuzzy edges is encountering the word often enough that you develop a feel for how people use it. (or in the case of りんご, living in Japan and encountering the real object often enough, maybe.) You could put the whole wikipedia article on 鉱物 on the back of your SRS card and you'd still have a mental concept which was a bit fuzzy round the edges. That was really all my point was -- that sometimes a long definition in either language isn't really going to add much extra value at the point where you first encounter a word and think "what does this mean?".

Quote:I don't know if using translation helps learn a word. Maybe it does. Maybe it's detrimental. But I don't think it's important. I think you know a word only if you can come up with typical situations where the word is used and natural sentences native speakers would say in such situations. If you can't, you don't know the word yet.

I agree with your definition of 'when you know a word'. But you don't put words in your SRS only when you already know them :-) The question is how you get from zero to 'learned' -- does method X seem likely to work? if I do Y might that leave me confused later? I think that the question of "does something help or is it detrimental to the learning process" *is* important, interesting and a popular topic on this forum...

(nest0r: yeah, I don't think we really need to have the critical-period argument again in this thread, do we? There's one in the archives somewhere if you want it :-))

I'm with Shirokuro, using a monolingual thesaurus is a very bad idea. I wouldn't even do that in my own native language. I use a thesaurus to find new words if I'm in a situation where I've overused a word or can't remember the synonym I want to use, but I would never use it to LEARN a word. That's more or less impossible, there's no such thing as a perfect synonym. If two words have identical meanings, one will eventually disappear or begin to differ in meaning.

Very good points from Magamo.

Why you shouldn't use a dictionary at all:
http://www.lingua.org.uk/voc.html
http://www.lingua.org.uk/vocdb.html