I did read (most of) the first post.

You seem to be saying your listening skills have not become very good. I think a big thing to get great listening skills is to use/input Japanese for long periods without any English. AJATT is not just about a large amount of input, it is about doing away with English and forcing your brain to adapt. Using English is in between Japanese input is like running to lose weight but stopping to eat a cup cake every 10 minutes. I have no idea if you are doing this, but it could be something that is holding you back.

As for the "listen over and over 40 times", I'm now going to disagree. Main reason based off anecdotal evidence: Karaoke. In Japan I sang karaoke a lot. Now, I know there are a group of 10 to 15 songs I've sung and listened to more than 40 times each. Yet, I could not tell you what some of the lines mean, despite being able to recite at length some of the lyrics from memory. Other parts I still can't sing without reading along.

I'm not sure what you are saying the problem is then. If you can recite the lyrics from memory then your listening skills are working perfectly. Not understanding the meaning would be more a problem of vocab wouldn't it?

@Thermal,

Let's be blunt: AJATT is a loaded term on these forums. People say "Do AJATT" as if there's a clean cut method laid out. There isn't. It was a method one guy poorly described in the early and study portions, but was detailed in the lifestyle section.

Early stages: Is it RTK or 4000 Hangul characters? What RTK Lite in all this? This forum handles this issue in detail offering information that can be tailored to your individual needs.

Study: What the hell is a "cool" sentence. How you do know a sentence is cool early on? How do we use sentences? What are good starting sources? This forum at least has provided a wealth of techniques and resources in this area.

Lifestyle and insights: Here we have plenty of tips and advice. This is where the AJATT blog shines.

Khatzumoto never detailed how much he studied a day. He sort blew it off as 'well, it's fun so it's not studying' description. So we don't know if he spent 1 or 3 or 6 hours of his free time on an average daily basis reviewing old cards and looking up words and sentences to put in his SRS. Something tells me in that 18 months, he put in 1500 hours worth of "study" that was enjoyable to him.

Now, yeah, I'd like to think that the listening 24 hours by itself would have helped, and I'm saying for me it did not. By playing comprehensible audio and reading comprehensible material and going to comprehensible websites and watching comprehensible movies I have more fun in Japanese. I detailed this 18 months ago when I said "Tiger and Dragon" was more enjoyable after having watched it once with English subtitles. A year after that, it was shows were more enjoyable if I watched them with Japanese subtitles. I want to get to the level where it's enjoyable without the need for subtitles, and I think I hit on it.

So yes, it's past the "I think this based on Khatzmoto's advice" and it's to the "I think this based on my experience". So, based on your experience over the months you've done this and the results you're getting, what is the advice?

Yeah, I'd like to think I could throw English out of my life, but not an option for a majority of working adults doing this. I'm limited to either a show or two at night on my laptop in my room or listening to the iPod on the speakers (which I'm now able to do at work).

About the listening, here's a personal story: Filipino guy that works at a Karaoke bar in Yokohama could sing fluently in Japanese "Pacific Hotel" and "Get Along Together". He's doing it from memory just by using romaji and listening to it a lot. He has no clue what he's saying. He can "hear" as a biological response, but the listening comprehension is zero.  When I say listen, I mean to understand what I'm hearing, not just being able to mimic it.

Plus, I posted this before: it wasn't increasing my vocabulary that doing this. I didn't add many new words doing this, but now I'm understanding words as they come out of the speakers.

Very true AJATT is a vague term.

Not saying what you are doing now isn't working, just thought I would add what has helped me a lot. I wasted 2 years using crappy study techniques and not getting very far and now I have reached a pretty high level, especially my receptive skills. Also thinking in Japanese is pretty much the norm for me now and occurs naturally.

I see what you are saying now. I guess part of it may be that you are not actively listening much? I think passive listening does help, but is no substitute for active.

I personally believe learning Japanese is kind of like training your muscles. It's when you are tired and keep working out that you get the real benefit. This is why English breaks allow your brain to recuperate and not become great at Japanese.

True life doesn't always allow for this. However, if you take out English, even if you don't have much time to do Japanese stuff I think the effect is almost the same. IE you brain has no choice to understand Japanese otherwise there is no internet. It's a sink or swim situation I guess.

Anyway, sorry for not knowing much about your situation and blathering on..

That sounds like a lot of work - it's a good method, but if you can keep that up for a few months it will be almost superhuman unless you're really enjoying doing things that way already.

I have found that the 80-20 rule that is mentioned on AJATT (pedaling vs. coasting) is a good way to frame it.  I make an effort to spend a couple hours each day SRSing and getting sentences, using the dictionary, etc... but after that, I just chill and go with it.  If I don't know a word, oh well - I just relax and enjoy what I do know.

Then again I live in Japan, so I don't necessarily have to make extra efforts to be surrounded by Japanese (putting it on while driving, etc)... but your method sounds a little bit strict!  Best of luck though, do post your results!

Right now I've been doing alot of listening and reading, but only a little writing really, typing is no problem though. After completing RTK in March this year I actually stopped studying kanji. I've gone on to sentence mining. I've been using iKnow and for vocab its actually working well. The best thing it does for me is bring new words to my attention. Since I'm living in Japan when I hear the word in the real world it helps reinforce learning it and usually I don't forget the word after I hear it in the real world.

I really want to try the srs2subs thing. I've hand picked sentences out of Okuribito and that seemed to work well, I can watch the movie and mostly understand it now without subtitles. I want to put stuff on my iPod too.

I think your method would work well, watch it one time through, next time go slowly and try to make sure you understand everything. Watch it again a third time and see how you feel, listen to audio without the visuals once you get a hold of the plot and how the story progresses.

I'd agree about the whole white noise thing. I've been listening to Japanese music for almost a year now and if I don't know the vocab for what they are saying its just white noise to me. Although after I learn the vocab I actually can hear and pick out the words in the music.

Another update on this front. First, the "use J-Drama scripts and J-Drama audio to make your own version of Audio Books" does work. It's really useful for shows that do not have subtitles readily available.

Second, I'm trying an experiment to carry the success of what subs2srs did to my passive listening and see if it works for reading as well. So here's my steps:

1. Take a drama script that I subs2srs (絶対彼氏 in this case) and print it out. I prefer 2 columns printout with 1/2 inch margins.
2. Go through, and mark the words and phrases I do not know in the non-dialogue portions (the dialogue I took care of during the subs2srs phase).
3. Enter these words into SRS if available or at least figure out the phrase.
4. Now the interesting bit, shuffle up the pages and read through the right column then go to the next page. After all pages are read, shuffle and this time read through the left columns.

This is simulating the shuffle function on my iPod that I do with J-Drama audio. One column of text takes me about 3 to 5 minutes to read which is about the time your mind might start wondering. After that, I'm onto another column from another part of the show or a different show. Obviously, I can only do this reading but it was remarkably effective.

I'll try this with all four dramas I currently have processed. Sadly, the last show I did I do not have a script for (but I do have the book). However, the next three shows lined up have scripts to add to this. In addition, I'll try to sneak in an unprocessed script as time goes on.

Benefit: Well, assuming I keep reading them, each page read is about 10 minutes of reading. Doing these a few times feels like it'll translate to help my reading like the audio portion did for my listening.

To be honest, I noticed I wasn't getting benefits of reading Japanese material (manga, news sites, blogs, etc) similar to not getting benefits from listening to lots of Japanese. This got me thinking I need to train my reading in a better way other than just adding more vocabulary and grammar. I'll keep you all updated.

Last edited by Nukemarine (2009 December 05, 3:03 am)

Tobberoth wrote:

Nukemarine wrote:

To be honest, I noticed I wasn't getting benefits of reading Japanese material (manga, news sites, blogs, etc) similar to not getting benefits from listening to lots of Japanese. This got me thinking I need to train my reading in a better way other than just adding more vocabulary and grammar. I'll keep you all updated.

I'm just waiting for the day when you come into this topic saying "I just noticed that using Japanese doesn't seem to benefit my Japanese. Therefor, I'm now going to study nothing but Russian. Hopefully, that will make my Japanese better."

I'm sorry Tobberoth, I thought you already knew the above just like you knew that listening to Japanese one did not comprehend would not be beneficial (that was your first response to this thread, no?).

By the way, I'm assuming that you're only joking, but I'm still doing a long reply.

I'm trying to figure out if there's a critical mass one can reach in all this. Truth be told, I'm thinking back to the time I was in five to eight years old. During those years, I distinctly remember reading and re-reading the same books over and over again. At some point though, it got to where I only wanted to read a book once. So maybe for my mind, in how it worked, needed to devour the simple stuff early on many times before it could digest the complicated stuff with no problems. Am I experiencing the same effect now with Japanese? Maybe.

All I'm doing in trying to adapt to problems as I see them come up. Posting the scripts around my room ain't cutting it, cause it's boring to change them around. Reading from the computer hasn't been cutting it. Nor is reading a novel or manga, as I guess I'm getting bored with the effort. So yeah, I'm stepping back to see if I can advance faster. It worked so well with listening, that I'm able to talk pretty much in Japanese with the guys on base. So now I find out if it can work with reading (which will in turn improve my speaking and writing).

And yes, this is just me. Others would be bored with the idea of processing an hour of drama, listening to it a lot, and reading its script a lot. They'd rather take a manga and sentence mine that. Another may just study vocabulary, grammar, and go to local bars and chit chat. We all have our ways, and they all probably work to a certain extent. I'm just recording how I'm doing it.

harhol wrote:

If you want to improve your reading skill like a native Japanese would, you could always try the Graded Reader series.

True, those are also options and they were tempting to me a couple of years ago. There's plenty of stuff out there to read, from news, to books to manga that cover all the age groups. I've read quite a bit (I think) so that wasn't the problem.

What I want though is a lot more bang for the buck. This might be the way. So now I get a lot of variety in little bite sized chunks. The reading equivalent of the "aural snacking" that Nestor made an observation about early on.

Plus, and it's a sad thing to admit, I hate sentence mining. I don't mind doing SRS, but the grabbing stuff (actually typing them) from books and putting them into an SRS just makes me want to stop. Didn't really like it early last year, and I still don't like it. Hell, I don't even like looking words up in dictionaries when I'm reading. Using subs2srs mixed with these scripts sort of give me the best of both worlds then. With this, the sentences are pretty much in the SRS, and I'm just looking up the rare word (well, 60 words over 11 pages) I didn't know from the narrative.

All said and done, I'm left with the equivalent of a chapter from a book that's at my level. My very own reading aid.

Warning, don't try to read this. ;p

To be honest Nukemarine, yours and my own recent posts and reading this book "Reading Japan Cool" (though I hate the title, hehe) has got me thinking more about this thread as well. Not just the sections about the notion of multiliteracy, reading outside of school, and the relation of manga to anime adaptations (and other branchings out into media), although I think this is relevant to the possibilities of subs2srs and the overall immersion ecology.

They also discuss 'rereading' (amorally Google/acquire it or see ch. 5 here) strategies for enjoyment and comprehension, rereading in a single or multiple sitting, and how developing Japanese manga readers process dialogue, faces, background, different aspects of the story rather than focusing on it as a whole, et cetera. It also mentions the differentiation in the culture of school literacy vs. outside reading communities (language education has a long way to go, no thanks to phonocentrism ;p), and self-study strategies additionally influenced (helped/hindered) by motivations of social currency and embarrassment.

This and the overall 'graphic' context esp. the 'iconic' aesthetics got me to thinking about the importance of understanding the layers of literacy, especially in cultures where logographs and visual iconicity have such contextual weight with regards to visual/semantic processing in the brain [related link] alongside syllabaries. We haven't done nearly enough balancing of this visual aspect of language literacy, especially in this day and age where kanji production is so 'mechanically' easy and we have new methods of self-study (thus you can look at, via RTK and similar, customizing and suborning the components that make up kanji not to 'production' but to 'recognition') [related link], and we're learning different strategies for integrating memory, rethinking notions of short and long-term as one develops working memory skills (keywords: retrieval structures, chunking, 'conceptual blending' it all together in the 'global [neuronal] workspace' - Also: here, here, here, and here)...

*See David R. Olson (ch. 4) or Florian Coulmas ("What is writing?") or Masako Hiraga (pp. 213-246), Roy Harris ('integrationism') for more on leveling the field when it comes to looking at phonographic and logographic systems without squashing down communication to a phonetic level (this influence hasn't done East Asia any favours -- well that's not fair, it did help redirect thinking towards democratic rather than elitist notions of literacy, but as Nakatani suggests, Derrida provided some nice counterbalancing**), or Ong/Havelock on 'secondary orality' vs. the essentially extinct 'primary oral' noetic ecology... I'm also interested in what notions of lexical priming and information value might say once the new methodologies are improved and applied to non-alphabetic writing systems, especially in relation to cognitive studies on the 'time course' of access (I believe glowingfaceman had an amusing blog entry tangent to this reference to time), the abilities of Mandarin/English children when it comes to recognizing characters/words in speech, deaf/memory differences in Japanese literate folks, the influence of this concept of imageability... Oops, meant to throw in some stuff about comprehension and skimming/decoding, autonomy/context, etc., but I guess it's implicit in the above (or see Hiraga on p. 225 or Ch. 7 of Reading Japan Cool)...

Ah, where was I. Oh yes, putting it all together with my previous obsession with using multiple senses strategically together, looking at how to combine SRS/non-SRS optimally, optimizing 'fun' [rel: affective learning and 'flow'] by centering the focus around personal preferences and real media, allowing oneself to 'fail in layers' (ie my defenses of i+x rather than i+1 with regards to multiple fail points being okay due to relearning and one's attention controlling automated algorithmic spacing of cards [SRS as heterophenomenological life partner]), and my recent strategies of breaking up what I subs2srs along more Nukemarinish lines, I am now interested in combining all this media together in a more integrated fashion, using the different sources for different layers and contexts but interconnected, designing them to be constructed and skimmed very easily yet consistently...

Anyway, unsurprisingly I've lost my train of thought, but this specifically relates to what Nukemarine was just writing about reading, and my own plans for incorporating reading... Must consult my notes, especially on subvocalization.... *wanders off muttering and sifting through .txts and papers*

Edit: I suppose right now I'm just planning ahead for criss-crossing written/video/aural sources via manga/light novels/BIJL/RRJ and subs2srs, eventually collapsing them and simply extracting extemporaneous selections. With regards to reading, I'm interested in how it is parlayed into a kind of mental textural element after blending visuospatial/semantic/phonetic/&c and abstracting it, so I think it's also important to be able to treat the mediums/senses asynchronously... *wanders off muttering again*

**I shouldn't cite Derrida's Of Grammatology too much because of his coding things in terms of ethnocentrism, even if effectively in terms of influence on modern linguistics it's somewhat true and he explained his understanding of nationalism not always being imperialism (as in the case of globalization analysis from within the endogeneity trap as Saskia Sassen would say -- or the belief in phonocentrism as the only proper democratizing doctrine), the universal nature of phonocentrism and its difference from logocentrism but anyway, Han-liang Chang, as well as Kojin Karatani and Atsuko Ueda have pointed out native origins to these thoughts privileging sound via the illusion of separation/primacy in literate/secondary oral cultures...

Last edited by nest0r (2009 December 05, 9:41 pm)

Although there's been a lot of changes to how I'm approaching reviewing in all areas, I'm just updating with regards to items mentioned in this thread.

First: Still no problem with using subs2srs get the best from my listening. I'm going to consider this a primary recommendation to anyone wanting to start Japanese.

Second: Reading dramanote scripts. The mixing up the pages to read random columns work well. It's been useful to do while I'm eating in the chowhall (US Navy term for a cafeteria) or having to time to kill. In addition for us professional workers, it looks less like goofing off when you're reading off printed sheets of paper than a manga or paperback book.

Third: I stopped adding/activating words in my vocabulary deck. From now on, I'll just go off definitions I added to the subs2srs cards. The main reason is I'm going to use my vocabulary deck in a more systemized fashion thanks to the sorted Core lists provided by Cangy.

Fourth: After my SRS vacation (started last month, will end when I leave Africa), I'm going to do subs2srs spread throughout vocabulary and grammar. Right now, I'm thinking two hours of shows converted, followed by 250 grammar sentences, then two more hours of shows to convert, followed by 1000 vocabulary sentences, repeat. The grammar and vocabulary are what's left of my "JLPT 2" equivalent material which can slide me into high level area.

Fifth: I'm going for simple and enjoyable dramas to subs2srs for now. Complicated dramas (5cm a Second, Around 40, Last Friends) are very enjoyable to watch but simple dramas (Zettai Kareshi, Gokusen, Hana Yori Dango) were not as draining on me when I did subs2srs with them.

I will say that I'll be glad when I'm back in the US and Japan if only for the broadband connection. It's good that I got to be creative with my learning methods in Africa, but I feel I'm not where I could be at without all the restrictions placed on me here.

In my opinion, reading Japanese texts (preferably with an audiobook) will give FAR more benefit than the other things, such as TV shows, video games, etc, because it's simply more exposure packed into the same amount of time. For me, sentence mining, subs2srs, etc really aren't enjoyable and makes you tired of the source material very quickly.

Last edited by PATRICKRL (2010 October 31, 10:40 am)

aberu wrote:

Wow.. I can't believe I missed that unbelievably important part. I have been listening to the same loop of 10 or so different, minute-long news articles in random order for about 2 weeks, and nothing has been sticking. I think the frequency between and speed of the articles has been so great that I am not retaining anything from them, all the while adding more articles to the playlist I listen to. That seems like common sense to continue to listen until it sticks, I just thought it would work itself out in the end if I listened to it enough times. I should SRS these sounds too. I am going to try cutting it down to just 2 or 3 and see what happens. Thanks aphasiac

Actually what I said there wasn't 100% true. Khatz definitely recommends relistening to the same audio over and over (he states he can still quote large blocks of The Matrix and certain Disney movies in Japanese), BUT he never says stick with one thing and stay with it. Cos that wouldn't be fun, and AJATT is all about fun.

Definitely a few passes of each audio item are needed; usis35's method earlier in this thread sounds best to me.