Rigabamboo here with another installment of the best of the AJATT+ forum.
We love to rattle on about MCDs on the forum. We’ve been at it ever since Khatzumoto introduced them to AJATT+ here. What is it about MCDs that causes so much discussion?
In a nutshell:
They’re simple.
They’re powerful.
There’s nothing like that “is it really OK to have so many cards this easy?” simplicity combined with that “holy cow I can now write a bunch of kanji compounds like a native” effectiveness.
But let me backup – what is an MCD anyway?
Apparently it stands for Massive-Context Cloze-Deletion Card.
I didn’t know that until I looked it up just now. Am I qualified to write this post? Most assuredly not. I’m a latecomer to the MCD craze, and I admit to being a bit confused about what the heck they are and how to make them.
So I made this forum post: “Show me your MCDs!” asking AJATT+ members to copy and paste their MCDs.
Here’s one アッシュ posted:
Front:
第二に、頭が柔軟であり新しいことの############収 が著しいこの時期にこそ外国語である英語のリスニングやスピーキングなどを学習する必要があるからだ。幼いうちは、年齢を重ねてからでは身につけることが 困難な部分さえ補うことができる。これに関しては、英語の学習を中学から始めて、留学などの特別な経験を経ていないほとんどの人が実感しているはずだ。
Back:
吸
++++++++++++++++++++++++++++++++++++++++
きゅう‐しゅう〔キフシウ〕【吸収】
［名］(スル)
２ 外から内に取り入れて自分のものにすること。「知識を―する」「大資本に―される」
第二に、頭が柔軟であり新しいことの吸収が著しいこの時期にこそ外国語である英語のリスニ ングやスピーキングなどを学習する必要があるからだ。幼いうちは、年齢を重ねてからでは身につけることが困難な部分さえ補うことができる。これに関して は、英語の学習を中学から始めて、留学などの特別な経験を経ていないほとんどの人が実感しているはずだ。
As you can see, there’s a lot there. That’s the “Massive Context” part.
The ############ is the “Cloze-Deletion” part. It’s hiding just one kanji: 吸.
You see the card, you produce 吸, and you give yourself a passing grade. The extra stuff is just to give you context for the missing kanji. It reminds you where you found this chunk of text and how the kanji being tested is used. It’s not there to be read – just glanced at until you recall the answer. You might not even have to use it at all, but it’s there.
So there’s a first taste of MCDs for you. This example is for more advanced players, and there are ways of making things even easier for total noobs. Anyway, more on that later…
Ive tried a different technique, and it works wonderfully.
Once I have gotten a grasp the meaning of a sentence, I would change the card and cloze delete the part where I think I would have problems producing in real-time conversation. I call this Fluid SRS (referring to the changeability of cards).
For example:
Front: あなたはどうして一人きりなのですか
Back: Why are you all by yourself?
Changes to:
Front: あなたはどうして一人___なのですか
Back: きり
It takes only a 10 seconds to change a card. The good thing about this is, once you tackled the first problem (understanding the sentence–according to the COMPREHENSIBLE input theory of Krashen–is probably the first thing we need to do), you can then shift your focus on other minute details of the language fact-sentence. You can focus on prepositions or compounds you may mess up. I don’t do this for every sentence–I do it only if I feel that I’ve owned the meaning of a sentence, feel like there is something I need to actively commit to memory, etc. You can always change the cards–if you feel that you’ve owned that compound or preposition, then you cloze delete something else that you notice you might have problems recalling in real time.
That actually gives me an idea for how to do my own cards. I’ve been intending to make sentence decks out of Nakama 1 and 2 (since I took Japanese classes in college long before I found AJATT, and the vocab and grammar in those books is the best link I have right now to my past, slightly-better-at-Japanese self), but I wasn’t sure how to put them in MCD format, since in the earlier stages of the book, vocab is all there is and it didn’t seem cloze-deletionable. But now, I think I got it:
Question side: Complete sentence with the vocab I want to learn, fully kanjified of course
Answer side: Reading of sentence, target word(s) and meaning
And once I feel like I’ve actually owned the card, I blank out the target word on the question side and leave the answer side as-is. It could just work…
Yeah. Sometimes, I don’t fully cloze the target word, beacause sometimes I just straight up forget. So I make it easier on myself by giving myself hints.
Another example:
Front: 真___はわかっていません
Back: 相
If you fully leave out the whole word, you will forget it. But seeing the 真 part, I immediately recall the 相 part. I always nail this card. It’s a weird phenonemon but there are manytimes where the I can say the complete compound of a partially clozed target word. Maybe it’s a mechanical thing, where I drilled it so much (by actually saying the sentence) that it just comes out with little to no thinking. And this is what I want because in a conversation, that’s how it should come out (learning language is more like tying a shoelace or riding a bike more than it is learning a historical fact). The beautiful thing is that you already know the meaning, the function of learning a particular sentence has changed–you just want to commit the target word/preposition to procedural memory systems so that it comes out naturally in conversation or writing. Another great thing about that is that you work on the meaning (only this time unconsciously, your conscious reticular activation system only allows you to focus on a few things at a particular time) so that you get to own the meaning even more.
I don’t like cloze deleting a sentence the first time, especially since I haven’t fully owned the meaning yet. If I have owned it, then yeah, it’s time to cloze delete. First time I tried cloze deleting, I could make out what the cloze deletion was, only I would forget what it would mean, and I would have to reformat the card into one where I memorize the meaning of the sentence.
Also, these days, I do this with cards where I memorize meaning of the sentence:
Front: 工場は生産（せいーさん）を縮小（しゅくーしょう）せざるをえなかった。
Back: The factory had to cut back its production.
I furigana-ize my sentences because it’s easier for me like that. It’s easier to just focus on the meanings and not worry so much about the readings. I think it is too much work to try to “memorize” both the meaning AND the readings. But if it works for you no problem, great!
I’m still quite new to MCDs and it took me along time to even try them out but what really helped me see how effective they can be was when 勝元 did a post on ajatt plus demonstrating MCDs using English. Maybe I can help some people still who are confused about MCD’s.
(Quoting from the above article)
【Apparently ############ stands for Massive-Context Cloze-Deletion Card.
I didn’t know that until I looked ############ up just now. Am I qualified to write this post? Most assuredly not. I’m a latecomer to the MCD craze, and I admit to being a bit confused about what the heck they are and how to make them.】
—————————————
Can you guess whats in the goes in the ############ ?
—————————————
A second one;
【Apparently itstands for Massive-Context Cloze-Deletion Card.
############ didn’t know that until ############ looked it up just now. Am ############ qualified to write this post? Most assuredly not. I’m a latecomer to the MCD craze, and ############ admit to being a bit confused about what the heck they are and how to make them.】
—————————————-
SO imagining your a native Japanese speaker learning English, who has gone monodic, this is how you would do MCD’s. So just reverse the process and make it Japanese. だろう？
I hope i have helped, like i said this really helped me understand how they can be helpful and how the ‘context’ part of MCDs really comes into play when guessing the missing part of the card….or maybe i have just confused even more of you =S.
p.s I know I haven’t given the answers, just scroll up to the article at the top and read to find the missing words.
Is it possible if someone can translate this to English or give an English example? I’m not learning Japanese so I don’t understand what just happened there haha. ^^;
Sure, I’ll do one for you:
Source: Lifehacker.com
Front: Tips and tools are useless if you don’t know the basics. If you’re still struggling with the ###### concepts of programming and design, we’ve got a few lessons that can help you. To get started with web development, our Lifehacker Night School series can teach you HTML and CSS as well as JavaScript, which pretty much covers the basics. We also have Photoshop lessons which can teach you how to mock up a web site layout. We also have a basic lesson on color theory. To pick up a few more design skills, check out this list of resources. It’ll help you get inspired and discover ways to learn more of the basics. If you want to learn more, check out Code Academy for additional programming lessons. When you’re ready to take your first coding project from start to finish, be sure to read this.
Back: initial
of, pertaining to, or occurring at the beginning; first: the initial step in a process.
Our initial admiration for their efficiency gave way to disgust about their methods
The initial stages of a syndrome may differ vastly from the final symptoms
Wow that sounds very useful. Thank you so much! 감사합니다~ ^^ I’ll thank you on behalf of all the non-Japanese learners as well haha. Oh and of course Khatzumoto as well.
(Taken from AJATT ‘about’ page)
Front:
Hey! Thanks for visiting! This site is about how ############ can learn Japanese without taking classes, by having fun and doing things ############ enjoy—watching movies, playing video games, reading comic books—you know: fun stuff! Stuff that ############ feel guilty about doing because ############ should be doing “serious things”.
Back:
You
Hey! Thanks for visiting! This site is about how you can learn Japanese without taking classes, by having fun and doing things you enjoy—watching movies, playing video games, reading comic books—you know: fun stuff! Stuff that you feel guilty about doing because you should be doing “serious things”.
Oh I almost missed this one. I thought you could only learn for words, but you gave can example about how you can use it in grammar. Thank you very much! ^^
—
I’ve already added some MCD cards but haven’t reviewed them yet. Excited to see how this works.
Here’s an example of a bilingual MCD I’m using to learn Spanish:
Front:
No tengo dinero […].
(I dont have enough money)
Back:
bastante
———-
tengo – i have
dinero – money
bastante – enough, a lot of
I usually cloze delete the same sentence several times, for example:
Front:
No […] dinero bastante.
I dont have enough money.
Back:
tengo
———–
tengo – i have
dinero – money
bastante – enough, a lot of
And one more thing, for a language like Spanish I will sometimes cloze delete 2 words (sometimes 3!) at once in order to teach syntax. For example, same sentence:
Front:
No tengo […].
I dont have enough money.
Back:
dinero bastante
This is useful for teaching syntax, and in this case it teaches me to place the adjective after the noun. These can be more difficult since you have to recall 2 words but are great for learning syntax.
@JustinB, forgive me but the sentence you just used rings wrong to my native spanish. I believe you meant to write, “No tengo bastante dinero.”
That or “No tengo dinero ni nada que dar”, but that could totally just be because I worked in a Mexican restaurant and heard a certain song over and over and over and over again. =p
And yes.. I do realize it’s saying something different, it just instantly came into my head as soon as I read the sentence.
hmm, perhaps my sources aren’t very reliable then? That’s kind of demotivating given that I retrieved the sentence from the first entry found here: www.spanishdict.com/translate/bastante
Thanks for pointing that out though, I think I might steer away from spanishdict.com from now on.
Actually, I’ve found that example sentences from bilingual sources tend to be…something to be avoided. Many times, the sentences are just not native quality, but it’s difficult to tell that early on. They’re still useful for getting an idea of what a certain word or expression means, but aren’t necessarily good for emulating.
So to all of the Japanese (and the rest of you!) learners obtaining sentences from Tanaka Corpus, Eijirou, tatoeba.org, et cetera, you might want to seek better (read: native) sources.
Could parallel texts work?
For learning Chinese what is your preferred method? Many words are two Hanzi compounds-Does it make more sense to test the cloze on one missing Hanzi of the word that is made up of two Hanzi or the whole word (2+ etc.. Hanzi)?
Making a cloze deletion with just one hanzi would be my recommendation.
I think Cloze deletion is an excellent method for learning, BUT I see some difficulty in its implementation. If you are a novice (or even intermediate in your knowledge of the language) you will cloze delete a word that is very hard to guess at given the background of the sentence. To avoid this, you will spend an undue amount of time trying to figure out what word is easy to guess given the background of the sentence. For this reason, I think that I will continue the tried and true sentences method, but perhaps cloze delete particularly stubborn words.
I think this is where liberal deletion comes in. It’s not difficult to figure out which cards have clozed words which are very hard to guess at even as a beginner – you’ll hate them! Just delete the ones you don’t like and you’ll be left with all the easy – but still useful – stuff.
This MCD thing seems awesome, but I feel like I’m the only guy here who doesn’t understand how you look up the meaning of these sentences in both the sentence method and MCD. I’ve read through almost all of AJATT (currently on Kanji stage), but I couldn’t find Khatz explain anywhere how you translate the target sentence into English for SRSing. It seems easy enough for a sentence like 今、何時ですか。where you can translate word by word and it makes sense, but for something more complicated, I don’t understand how you get the English meaning. Any help would be very appreciated.
Your post brings up an important point that may be easy to miss when reading AJATT. In the bilingual phase, the sentences should always be professionally translated. The most common, and easiest, source of these sentences is a bilingual dictionary. Khatz also has a bunch of sentence packs available for a reasonable price.
In the very beginning phases I used example sentences from Spanish-English dictionaries. For example:
Front: […] lo sé.
I already know.
Back: Ya lo sé.
I already know.
Front: Ya lo […].
I already know.
Back: Ya lo sé.
I already know.
Sometimes I would include an English definition of a word on the back of the card. Eventually I moved into partial bilignual cards. I drew sentences from Spanish-English dictionaries and from Spanish books, websites, etc.
Front: […] tela seda rojo.
Back: Una tela seda rojo
seda = silk
Now I am using full monolingual cards with a monolingual front of any length I want, and the back is the complete definition of the word from a Spanish monolingual dictionary.
Thanks for the clear response. This definitely makes me feel a lot better about what’s to come with sentences. I was starting to get nervous (making it harder to keep motivated) about how sentences would work, and this alleviates a lot of those concerns. I really think Khatz should add something about this in the FAQ section because there’s no way I’m the only one who missed that. Sorry about the double post by the way. I didn’t realize that there was a lag between when you post a comment and when it appears on the site.
No problem. Just keep your head in the phase you’re in. I highly recommend the Quick Reference Guide (QRG) and the MCD pack from Khatz. I have both and they helped clear a lot of things up. The benefits are worth several times their cost, and you can get a full refund if you don’t like it. If you were to only get one, I recommend the QRG for giving an excellent macroview of AJATT.
MCD seems like an awesome technique which I intend to employ alongside sentences, but I feel like I’m the only guy here who doesn’t understand how to look up the meanings of all these authentic sentences you want to SRS. I’ve read through almost all of AJATT (currently on the kanji stage), but I couldn’t find any specifics about how to “translate” these sentences you mine. It seems easy enough for something like 今、何時ですか。where you can just look up each word individually and the meaning is pretty clear, but for something complicated, I don’t understand how to find the english meaning of your sentence. If anyone could explain this to me, I’d really appreciate it.
@Matt
I wouldn’t worry so much about finding translations of the sentences. Once you finish the kanji, you will be able to understand mostly what the sentence is talking about. This is where kanji is your friend, the more you know the better. Just keep going and your brain will start to make sense of it all.
I can’t answer your question with experience as I went through the beginner – mid intermediate stages with mostly pre-made decks(regret that now), but it seems to me that you just need to make sure that the sentence is truly i+1. If a sentence feels really complex, there’s probably more than one thing in there you don’t have a grasp on. So, you ditch it and go for something closer to what you can confidently understand. You can come back to it later, if you like. If it was worth learning in the first place, you’re bound to run into whatever tripped you up again in the future anyway.
If you’re sure that it’s just one thing you’re having trouble with, I recommend searching for it in various dictionaries, both J-J and J-E, and reading a bunch of example sentences and some differently worded definitions until you really feel you’ve got it. It’s rare that this didn’t work for me, but if it didn’t I just gave up on the word entirely. I would invariably run in to it at some point later on and be like, “Wow, I remember stressing over that word/grammar/etc. What the hell was I even confused about? It’s so simple!”
Anyway, that’s my take on it. You’ll run into places where you feel like the gap between what you know and whatever you feel is the next ‘step’ is just too large to bridge with i+1, but keep at it and you’ll find the sentences.
That makes a lot of sense. This in addition to professionally translated sentences seems like the way to go. Thanks for the advice!
I’ve always found it hard to remember what most podcasts teach because of the lack of review between lessons. But after learning about cloze deletions, I’ve been trying them out with some podcast dialogues.
It’s helpful for learning short phrases and sentences or grammar points. I especially like to use this method with question and answer sentences.