So I'll explain my view of language and translation and, basing on the view, answer the remaining questions.

First of all, I don't think most of basic words have meanings per se. Can you define every simple word you know like dictionaries do? I doubt that. At least I can't define all the simple Japanese words in spite of the fact that it's my mother tongue. English words are even harder for me to do that.

In my view, words are triggers to bring back memories. When you see/hear a word, say, "cat," do you automatically define it? I think what would occur in your mind is a kind of flashback. I think your brain searches for all the memories associated with the sound/spelling of "cat," and, if you focus on a single word, extracts an abstract image that is sort of the largest common denominator of brought back memories. If there is such a thing as the "meaning" of the word cat at all, I think it should be a verbalized version of this abstract image.

Words can be connected by certain rules and form a sentence. Again, I don't think a sentence has a meaning in the conventional sense. I think it's a sequence of memory-triggers, or words if you will, aligned such that a human brain can watch a "movie" of which each frame consists of memories triggered by a word in it. Word order, intonation, and other aspects of language found in a sentence are, I think, devices to create a story by smoothly adjoining seemingly random frames. In other words, grammar and such is the mutual agreement between you and a person who speaks the same language. They are a standard of protocol of story telling. Grammar narrates while your brain is experiencing a sequence of flashbacks.

If a sentence contains a word you have never heard or read, you can't see the corresponding frame because no memory will be recalled. The trigger fails to go off. But if other triggers hit your memories and if your brain can decode the rest of the movie by following the protocol, you might be able to guess what the frame you missed was supposed to be like.

Of course, there are a lot of rough points in the above theory. For example, I didn't define what a word is. An article is a word? How about a function word like particles? There are many other problems, but the purpose of this post isn't to give a scientifically rigorous model. I think this rough explanation suffices.

Translation is, therefore, swapping of memory triggers and reordering such that the encoded movie is considered similar, close, equivalent or faithful, or anything the person who re-encodes thinks is a translation. Grammars of two different languages might have similarity if human brains have a certain innate circuit to govern protocol to project a memory movie. However, if the two languages have developed independently, it is highly unlikely that there are a pair of memory triggers of the two tongues that bring nearly the same memories in most of the people of both languages. For this reason, I think translation is never faithful or equal to the source. I venture to guess that translation between two languages that are culturally different to the extent that speakers of respective languages have very different back grounds, experiences, and memories is asking for the impossible. I think any translated sentence between English and Japanese is a made up sham.

I don't know what the best method to acquire a language, and I do think anything is good as long as the learner is satisfied with his or her progress. So, if you're making progress, then your noun-first approach should be good. I don't know if every natural language has a grammatical object that falls into the noun category. Probably noun-less languages are extremely rare, I guess. Fortunately Japanese has nouns, though what happens in native speakers minds when treating a "noun" seems to be a little different. Anyway, I think your approach may work.

Tangentially related is difference between two supposedly equivalent nouns. This isn't important at all, but if you didn't know, even simple nouns can't be equal. Apple and りんご are different, for example. As is often said, the former is smaller, and connotations are different for obvious reasons; two different cultures will unlikely assign exactly the same implications. Water and みず are different for the same reason. I think you already know that みず can't be hot while hot water is ok, but it seems that the difference is larger than that. If you look at the two nouns as memory triggers, みず and water call very different common denominators of memories, at least to my mind. This might be obvious because at least memories associated with hot water has absolutely nothing to do with みず.

I already built a wall of text, so I will cut to the chase: I believe that learning a language is experiencing the world in the target language so that memory-triggers, i.e., words, can go off and bring back the actual memories and that the brain develops a circuit to handle the new protocol.

I don't know if we can "borrow" part of memories you have in another language, but this is highly unlikely because there is no way your brain can tell what kind of memory the average native speaker of the target language will recall with a given word. Your conscious mind may be able to learn the new culture, but it doesn't seem that the memory system and language circuits in your brain work in conjunction with your conscious mind. For this reason, I guess anyone who says a person can reach native fluency in a few years is wrong, lying or deluded, or their definition of native fluency is quite different from mine. He or she might be able to pass for an educated adult native speaker if he or she is good at faking fluency, but I guess it's impossible to experience the world in a few years to the extent that a sufficient amount of words can recall memories to the same degree of the average educated native speaker.

Obviously my experience in English is so shallow that words can only recall a tiny portion of my life. The depth of memories a Japanese word can bring back is tens of thousands times deeper. I said I may not be able to be fluent like a 10 year old in one of earlier posts, and that was in this sense.

SRS a textbook such as Tae Kim? It'll be a nice idea if you want to get a bunch of mnemonics. But language learning-wise you get nearly zero experience. No memory will be associated to project a movie in your mind. You can get a ton of experience to learn what a sterilized textbook-ish sentence is like, though. But then again, it seems some people find SRSing textbooks very useful, and I think it can be if you're an absolute beginner.

Translation? It could help grasp the senses of words and phrases, so it might enhance your language experience in the new language a little bit, though the newly stored memories will be inevitably skewed by the influence of the language you're basing on.

Anyway, this is why I think immersion is the most essential part of language learning.

I don't know much about philosophy, so probably I'm wrong. But there are a couple things I can't quite agree with.

"Desert" in your mind and that in an Australian living in a desert should be quite different. Of course the advanced communications technology has helped us to fill the gap. TVs, phones, internet, books, etc. are all helping us to share knowledge. But seriously, do you really think "summer" in your mind and that in an Australian guy are basically the same? They celebrate Christmas in a hot winter day. Of course Australians consume tons of stuff made in the Northern Hemisphere, but still I guess "summer" has different connotations in Australia. I don't believe that the English language can magically fill that huge gap. Just because you can communicate doesn't mean you have a similar idea about the word. Or does winter naturally turn into a hot season in your mind if you speak English with Australians? It sure doesn't happen in Japanese. I know the fact that it's cold in August in the Southern Hemisphere, of course. But my mind doesn't work that logically. If a person says 冬, then my unconscious mind just takes it as a cold season regardless of whether he is from the Southern or Northern Hemisphere. My conscious mind would correct it based on knowledge, and maybe my brain will search for memories about 冬 used in the context of "hot season." So I can communicate with a person from the Southern Hemisphere without much trouble. But this doesn't seem to be a counter-example or anything. Rather, it supports my idea.

Another point I can't agree with is about translation. You'll be surprised how different a pair of "equivalent" words are if you get better at Japanese. For instance, 雷 (かみなり) and "thunder" are almost "interchangeable" between the two languages, but what pops up in my mind is really different. It's almost impossible to explain this, but it seems to me that Japanese and English see the world differently and classify things in radically different ways, so I don't think "concept" you speak of helps much. Maybe, better than nothing. But it could be harmful at the same time. It's kind of famous that "thunder" + "lightning" is sort of 雷, and this may be similar to the difference I mentioned between water and 水; water is お湯 + 水 in some sense. But the difference between 雷 and thunder is way more complected than that. Can you guess why English students in Japan have trouble understanding that "thunders" and "lightnings" are strange in a normal context? Can you guess why they do not say "waters" as often?

@IceCream

You can't skip this post because connotation and stuff isn't important at all for your study. I'll answer your question about my cards and SRS in the next post.

I'm still not convinced that denotation plays an important role in language. I believe that denotation arises only when a person consciously verbalizes an idea that pops up in his or her mind and that verbalization isn't part of everyday language communications because it doesn't seem that we consciously verbalize ideas when speaking a language in a normal situation; normally speaking is an automatic, unconscious process.

As for the word "desert," it might help us understand what we are talking about if we imagine, say, a little kid who saw a desert in her favorite cartoon. Assume that her language experience about "desert" is very little, which is probably the case with most of the British kids. The cartoon depicted the desert as a very fascinating place where she can have a lot of fun. She learned that desert is a hot but fascinating place covered with sand. That's pretty much all her knowledge about desert. She doesn't even know how to spell the word. She has no idea how hot it is. She doesn't know how cold it is at night. Because the cartoon showed a skewed image of "desert," let us assume that she is fascinated by the ideal "desert" and wants to live in the exotic, fantastic place.

A relative of hers visits her home. He happens to be an Australian guy living in a desert. He knows everything about desert. It's not a fun place where his grand daughter can fly with a magical flying carpet. Weather is severe and extreme. It's not a place where a little Brit girl can survive by herself.

She finds out her grand pa came from a desert. Now she demands that he take her to the desert. He tries to explain how severe it is to live in a desert. He goes on and on and on, and she doesn't listen. She just keeps talking about "desert" based on her little knowledge, which of course is the only thing she can do. No one can talk her out of it because she doesn't know a real desert.

So, obviously she can talk about "desert" with an Australian guy from a desert even if she has a very different image of "desert." It doesn't seem that her mind has a definition of "desert" based on denotation. She can't even spell the word.

But if he tells a few stories that depict the severe reality of desert, she may change her mind. Why? I think that's because she learned about "desert" from experience. Denotation will never dissuade her. Only connotation directly affects her definition of desert. Of course, she might change her definition when she learns a denotation if it brings her a vivid image. She will change her definition if she thinks "This book says it's a very scary place!," which is a new connotation. Denotation-ish definitions found in a dictionary or an encyclopedia will help her gain some knowledge, but language-wise those definitions can only give her new connotations. It's the experience she has with a tome that changes what "desert" means to her, not the definition itself. When she reads about "desert", she first pictures it in her mind, and then changes her "desert." Not the other way around. It's impossible.

It's often said that a man convinced against his will is of the same opinion still. I think it's true, and probably the reason is that verbalized persuasion seldom if ever evokes strong enough of a image in his mind. If memory-triggers recall very different memories, it stands to reason that changing a person's opinion isn't the easiest thing. You need a "shocking" story, video or picture to completely change what a particular word means to him.

I see this happen everywhere. You can't easily talk TheTrueBlue out of it because his experience is very different from yours when it comes to relationships. I don't think his strong images of girls and relationships are based on denotation. Judging from his writing style, apparently he relies heavily on denotation when it comes to knowledge. But it doesn't seem that his definition of "relationship" is based on denotation. He knows a definition. But it doesn't matter because its just a piece of "knowledge," which I think has nothing to do with "meaning" in terms of natural languages. When he says the word "relationship," he means a relationship in his mind, not yours. When he learns about your definition of the same word, he may re-define his by picturing a relationship that matches your description. Again, denotation isn't changing what "relationship" means to him. It's the picture in his mind that can change his definition.

So far we have seen:

1. A Brit girl can communicate in English and talk about "desert" with an Australian living in a desert even if she doesn't have any denotation about it and it means a different thing to her,

2. Denotation doesn't directly change what a word means to a person,

3. Connotation changes the meaning of a word, regardless of whether it's from a real experience or imaginary experience evoked by denotation.

I think you rely on denotation only when you consciously try to verbalize your thought. And I don't think you consciously verbalize everything in everyday conversations. If anything, it's rare. I think conscious verbalization is a skill that can be improved by learning grammar and memorizing definitions. You could say it's also part of language. Even so I think it's a very superficial aspect. If one wants to learn a language, I think it's the last thing he should focus on.

Translation may help you develop connotation in a foreign language, though it's inevitably skewed. Grammatical knowledge help you "decipher" a foreign language, which can evoke images, and hence it definitely helps you develop connotation. Grammar and denotation undoubtedly improve a superficial skill in processing a language.

This is why I think that translation, grammar, and denotation are merely learning aids or useful tools at best. I'm not very interested in the superficial skill in processing a language by my conscious mind, though I want to learn it later. What I want to learn is the language itself.

I'm not saying I need to learn the exact same images of words as your average native speakers. What I need is my own images that are deep enough. They may be very different from yours. I just need to experience the world in my second language to deepen my language. If I can borrow memories in my first language, then it may not take that long. But translation doesn't allow me to do that. Recalling past experiences while pondering in my second language might be able to do that. But then again, sometimes I forget which language I'm speaking, so...

IceCream Wrote:Anyway, i wanted to ask about your cards that you set up. Like with the "what are you getting at" example, that's loads more complex than the cards that i have.

It's loads more complex probably because I had already learned English. As you already know, I learned basic English grammar and the most important words at school like everyone else in Japan, though my English was as abysmal as your average Japanese guy. I also did semi-immersion before starting SRSing. Actually I had to; no information is available in Japanese when it comes to sentence mining and stuff. I waded through the AJATT blog, Anki/Mnemosyne websites and whatnot using dictionaries. If Katzumoto's Japanese wasn't great, I would have fallen by the wayside before starting to SRS. Reading about immersion into fun stuff, I went full-fledged immersion, and then started using SRS.

Also, I didn't want to put simplified sentences or segments like clauses because of my view of language. I can't get enough context from components. I don't say they're useless, but grammar points and whatnot are not that important to me.

IceCream Wrote:How did you mark the cards to begin with? If you couldn't understand the back, how did you get to understand the meaning, and what did you do with the card until the meaning had become clear?

By "mark," do you mean grade a card when reviewing? If so, since I don't care much about definitions and their accuracy/precision, marking a card isn't based on whether I can grammatically decipher the content or translate its semantics. I think I already talked about this, but I grade a card according to the depth of connotations/context I can remember. So if I can remember the "feel" of sentences and the context, then is ok. So what's on the back side isn't an answer. It's an aid taken from dictionaries and other sources to grasp the senses.

That said, lots of early simple cards were reviewed based on grammatical understanding too, especially when I thought grammar worked well as good memory hooks. So relying on translation/grammar could also be nice if you really need them, I guess. Probably absolute beginners can learn faster that way. It's just those learned sentences can't be used when you produce them; I don't want to use a phrase just because logic behind the expression suggests it's ok to use it. I can use a new expression more confidently if I use it because I'm in a situation similar to the one I found the phrase in; the biggest difference is that I'm the one who's going to say the phrase. Certainly I still often use SRSed phrases in the "wrong" situations, but it seems you can avoid a lot of common errors among foreigners this way.

Anyway, when I started SRSing, sometimes I got the wrong "feel" when I ran into a sentence for the first time and marked cards based on the wrong "feel," but I didn't care because most of the sentences I put on early cards were so frequent/common I would come across sooner or later and refine my own definitions, which were not explicitly written on the cards.

To avoid learning the wrong impression, I often:

1. tried to understand the definitions in my dictionary,
2. tried to understand the example sentences in my dictionary,
3. googled the expression in question to get more context and,
4. avoided mining a sentence when context isn't rich enough.

So it's impossible to SRS a sentence from a textbook like Tae Kim because it gives nearly zero context and I can't guess the semantics or grade the card. The only exception would be when I SRS a sentence as a typical textbook-ish example. In that case I'd be able to grade the card based on whether I can feel the textbook-ishness, though I don't think I'd mine a sentence from textbook.

Oh, actually I've mined sentences from Urban Dictionary when I looked up a phrase I met in context and found more interesting sentences in it. In this case, the sentences from UD are put on either the front or back of the card. If I want to remember the hilarity, it'll be on the front side. Either way, I put both the original sentence I encountered and funny examples from UD. I don't know if it counts as a textbook, though.

As for how to get meaning, I don't care much about denotation because I don't think it's the meaning. It's good to learn fine definitions, but it's not the most important thing. I've been using SRS to prevent forgetting a specific context/connotation I want to remember. As I already said, the point is that just because logic suggests it's ok to use a phrase doesn't mean it sounds natural. It should be used in a similar situation. The semantics deduced from grammar doesn't help much in this regard.

Also, I only picked up sentences I really liked, otherwise I would have had to put into SRS pretty much every sentence I came across. So most of the time, I didn't need to worry if my guess was right on the money. I already liked it. I don't like things that make me worry. You might want to pick sentences from context-rich stuff you love. That way, you have more sentences that you understand. I guess dramas, anime, manga and whatnot are very good. If you think translation-based learning works for you, then you can take advantage of professional translation too if your favorites are subbed/dubbed. Then again, different methods work for different people. Textbook examples are all proper and standard in a sense, so I wouldn't be surprised if some people found them more helpful than real Japanese spoken by native speakers. Besides, it's not surprising that beginners need certain guide.

IceCream Wrote:Oh yeah.... one more thing... just how powerful is the SRS in your opinion, in production? Do you think i shouldn't really SRS sentences said by men? Will it have a negative effect when i want to talk? i don't want to sound like a man!

I think it's very powerful for remembering vocabulary words and such. Usually I try to avoid using grammar when producing my own sentences, though sometimes I have to resort to grammar rules. So I don't know how it would work when used for grammar/translation-based production. Probably it's super powerful. But be warned: machine translators know much more grammar and vocabulary than you. SRS might greatly improve your grammar and vocabulary, but still you may not be able to do any better than them.

Here is my view about "You'd sound like XXX if you learned Japanese from YYY":
http://forum.koohii.com/showthread.php?p...3#pid63993

I don't know if this could help, but I'll post another example of my early cards here:

----------------------------------------------------------------------------------------
[Front Side] Yuka: Your guess is as good as mine.

What does this phrase mean?
----------------------------------------------------------------------------------------
[Back Side] I don't know.

Illustrative Sentences:
Jane: Are there any good movies playing tonight?
Alan: Your guess is as good as mine.

Jill: How long should we bake this pie?
Jane: Your guess is as good as mine.

If you want to know why she left me, well, your guess is as good as mine.
----------------------------------------------------------------------------------------

The example sentences on the backside was difficult. The original phrase was taken from anime (actually it's Elfen Lied). Because you have all the information such as the situation she said the phrase in, her tone of voice, the rough equivalent "I don't know," etc., its basic sense is crystal clear. The difficult points I couldn't understand when I created this card were:

Why does Jane say "any"?
Why does the last example begin with "If"?
etc.

LOL! デカポストね。頭がいたい。

IceCream Wrote:Look again at what you did with your cards. Would it matter if you had written 分からない on the back of your cards instead of "i don't know", or "well, you tell me"? Neither really provides a better definition than the other, and neither have the same connotation really as "your guess is as good as mine". But i guess the english version does help you think in english, so...

I use a card mainly to remember the encounter with a phrase. I don't rely too much on the information to understand meanings. It's just another memory hook that reminds me of context. I said grammar rules can be good mnemonics if they give rough meanings of sentences. The information on the back of a card is working the exact same way. I take definitions and such with a huge grain of salt, so if they say "X means Y," that means I get another mnemonic that may or may not be a good approximation of the phrase in question. I may use this information as a certain reference point to understand the phrase X, but I should keep immersing myself into the target language so that I can refine my connotations and get rid of the wrong associations.

That said, it's just a general principle and I have been using the information on the back of cards in various ways. Actually I relied on definitions and grammar more when I started using SRS, and as I get used to English, my learning method has been getting more and more connotation-based.

Anyway, because I think the meaning of a word/phrase basically consists of connotations, I don't think a card teaches me the meaning of a word in the conventional sense. A card can never contain enough information.

As for the denotation thing, I think I understand what you mean by "denotation," "concept," and similar terms. I guess it's like a certain core of the meaning of a word. I think this core thing is roughly the same as the universal sense, concept, etc. you're talking about.

I briefly mentioned "the largest common denominator" of connotations in an earlier post. I guess this is the core, or the concept/extension in some sense. As you get more connotations of a word, I think your brain will notice certain similarity between them. For example, if a word is almost always used negatively, then you unconsciously infer that the word has a negative implication. If the word is almost always used when the speaker is referring to a person, you infer that it's a word that only applies to a person. I think that this kind of implication, i.e., "almost always used this way" senses form the cores of words/phrases. You could say it's the "average" or "common point" of connotations you have.

When you're asked to define a word or how it is used, I think you say it to yourself and recall your memories associated with the word so that you can "feel" the sense of the word. When you say a definition of a word in a dictionary is "well thought-out," probably you mean that it matches up with those connotations that pop up in your mind. So I guess an excellent definition that everyone thinks is well thought-out is a very good approximation of the core of a word because it's kind of a true statement that holds for everyone's every connotation of the word.

You can form a new, important connotation by picturing, say, "desert" by reading a definition. Since you intuitively know that the definition should roughly describe most of the connotations the average person has about "desert," you may set the image you pictured in your mind as a reference point or the center of the meaning of your "desert." And then you will polish your image through experience. In other words, you can learn a word by its definition and then internalize its concept. As you develop connotations of "desert," you might find the definition you set as a reference point is slightly different from the average of connotation. You might also find new connotations give more information about the word. When these things happen, you'll unconsciously (sometimes consciously) refine the reference point so that it roughly describes most of the connotations you have, i.e., it's the center of the core you have developed.

The core of a word you know and my core of the same word would be similar in some sense because a core is the common characters among connotations. The more you gain memories associated with a word, the finer it gets. So people with similar backgrounds have very similar cores. If your life has been radically different from mine, then you and I have less similar cores. Either way, there must be certain similarity between yours and mine. For example, you and an Australian living in a desert have different cores, but a desert is a desert, so both of you should know that it's hot and covered with sand if you're a normal person. So in some sense there should the core of the core of connotations. I guess this core of the core makes it possible for you and the Australian to talk about desert.

The core of a word and the core of the core are still abstract images. But you may be able to verbalize it by using your conscious mind. In my view, a verbalized version of your core of a word forms your dictionary-type definition of the word. The core of the core is the one that makes it possible to communicate with people with different backgrounds. If a definition found in a dictionary is very close to the verbalized version of your core, you think it's a good definition, and people with very different backgrounds might disagree with you. If it's accepted by many, that means it's a good approximation of the core of the core.

This idea still makes sense to me. But then again, I don't know philosophy and stuff, so I could be completely wrong.