5 ms·
The Heisig method, which recursively breaks down Chinese characters into patterns with arbitrary meanings, can help you sidestep this problem. You're never deal
by jim-jim-jim 2y ago
The Heisig method, which recursively breaks down Chinese characters into patterns with arbitrary meanings, can help you sidestep this problem. You're never dealing with shapes anymore, but rather reconstructing stories from these stroke/meaning pairs. Since patterns consist of subpatterns, you can tweak the level of granularity until a sensible narrative emerges. Just recite that story as you move your pen.
It's a lifesaver as an adult foreign learner, but I don't really see anything preventing native writers of Chinese and Japanese from benefitting from this general process as well. I've wondered if the guys who pass those truly insane 6,000+ character exams have to fall back on some sort of hack at that point.
- eloisius 2y agoAs another adult Chinese learner, does something like the Heisig method really help with language acquisition or just memorizing characters? I’m skeptical because of the immense amount of time it takes to learn even without elaborate story construction for each character. I’ve kind of resigned to being a word processor idiot, and only memorizing characters in handwriting as a bi-product of usage.
- mbivert 2y agoIt may be anecdotal, but I once involuntarily trained my memory by trying to recall what I've been doing each day. After a few weeks, my memory noticeably improved. That's to say, the task may feel insurmontable at first, but if you give time to your body to adjust, it should become easier.
- graeme 2y agoThat's very interesting. Did the memory boost persist? If so do you do any maintenance exercises like the original ones or have you noticed other effects?
- mbivert 2y agoIt's difficult to say: I think shortly after this "incident", I started doing more maths/physics on my free time, which must have help this boost to persist. Thinking about it, this recalls me of a Leonardo quote (pertaining visual memory then): > « I myself have proved it to be of no small use, when in bed in the dark, to recall in fancy the external details of forms previously studied, or other noteworthy things conceived by subtle speculation; and this is certainly an admirable exercise, and useful for impressing things on the memory. » I remember reading about similar observations regarding visual memory, where students trained to memorize visual information would outperform their peers (observations perhaps in the 1800s/early 1900s, IIRC a woman was in charge of this).
- adastra22 2y agoYou've got it backwards. The Heisig method is faster, and less work overall. It takes about 10 seconds or so to set an image in your mind of the scene for a character, then a handful of reviews over the following weeks. Then you never forget it. Classical methods would have you drilling characters for hours upon hours of wasted time.
- eloisius 2y agoI mean to say that my current method mostly omits hand writing. I can use a keyboard or phone to write and I can recognize characters fine. But on top of just learning a language, is Heisig so effective that I will be able to also memorize handwriting each character? Or do people measure how useful it is by being able to memorize the strokes for many characters, yet fail to become fluent in the language otherwise?
- adastra22 2y agoHeisig will have you learn handwriting, yes. Because that is explicitly all you practice in that method. Aside: Do you learn simplified or traditional? I learned traditional. I would have anyway because my wife is Taiwanese, but I advocate others to do the same because it is arguably the same difficulty if not easier. And going traditional -> simplified is tractable whereas the reverse is not. Learning the Heisig method is similar: learning from a perspective of handwriting is easier, and you get the ability to read “for free.” It’s a better approach, even if you never need to write by hand.
- eloisius 2y agoAlso traditional, and I agree it’s easier to remember them than simplified characters even if they are more complicated to write.
- adastra22 2y agoSo you have a reference point for this. When you learn a character by the Heisig method, you go from meaning -> writing. You don’t bother practicing reading -> meaning. It turns out that it’s very easy to go from writing to reading, much like going from traditional to simplified, but the reverse not so much.
- throwaway2037 2y ago> language acquisition or just memorizing characters What is the difference, in the case of Chinese?
- eloisius 2y agoCharacters are not words. Most words are 2- or 4-character combinations. Individually they have fragments of meaning, so you may get a hint, but not enough to understand for sure unless you’ve learned the word.
- NoInkling 2y agoIn addition, the Heisig method doesn't teach "meanings" at all, rather "keywords", which in many cases are only kinda sorta vaguely related to their respective characters' (fragment of) meaning. Either way, they're mainly useful as an "anchor" or name to refer to the character by, and in many cases might as well be arbitrary. The real purpose is just to turn an initially indecipherable blob that is a character into something recognizable (and writable, if you care about that). In general, you're not learning meaning to a useful degree until you start seeing or hearing (when you can associate pronunciation) those characters in context, as you imply. Yeah sure, with Heisig alone you'll be able to see 鸡 and know that it (most likely) refers to chicken, but characters like that are a relatively small proportion.
- jim-jim-jim 2y agofwiw I studied Japanese, but I believe most of this still applies. It's divide and conquer. When you are reading the book, you are indeed just learning the characters. It's a significant ~2-3 month investment that maybe doesn't make sense unless you plan on living and working in the country. But once you've gotten through it, it absolutely feeds back into vocab acquisition, since the characters are now completely unambiguous to you. Much like how Latin/Greek helps with English, you can also work out what entirely new words might mean if you are familiar with their characters.
- latentsea 2y agoI did RTK. I also learned to read around 3k kanji. Turns out it wasn't at all necessary to learn to write that additional 1k Kanji in order to become able to read / distinguish it. The time is better invested in simply studying how to distinguish visually similar characters. That alone solves the problem directly.
- jim-jim-jim 2y agoI also did all three volumes and found the extra 1,000 to be a waste. Really polluted my Anki. > The time is better invested in simply studying how to distinguish visually similar characters. But you still have to know what you're distinguishing between, which might only arise after repeated mistakes. Heading off this frustration directly by studying characters may not have been the best use of my time in absolute terms, but it did wonders for my overall motivation and made me feel like I was doing more than treading water. Pre-Heisig I was reading specific books intended for foreign learners, while afterwards I was just reading the newspaper.
- latentsea 2y agoIt's somewhat a shame there isn't heavily curated Anki decks for doing what I call "disambiguation study" where you focus on cards that help you distinguish similar things from one another. It'd really speed things up. >But you still have to know what you're distinguishing between, which might only arise after repeated mistakes. I'm learning Korean at the moment and it's particular brutal for this IMHO. Some words have taken a long time to properly understand due to repeatedly mistaking them for very similar words, and there are a lot of these in Korean.
- latentsea 2y agoI did Heisig's Remembering The Kanji as my first step on my Japanese learning journey. It helped make the written language feel more accessible, and I took it on as a fun challenge, but in hindsight only learning to read is important and learning to write is mostly a waste of time. Learning to read is vital, but being a word processor idiot is how everyone is living their lives for the most part.
- Tor3 2y agoAs someone who switched to using a keyboard for writing at the age of 14.. yes. But at least I can write my own language by hand, if I have to, even though it looks horrible (always did). I can draw it instead of writing it, and it looks better (filling in forms and the like). For Japanese I use a keyboard.. should be fine but.. no. Whenever we're at a table and need to jot something down on paper, or near a whiteboard, I feel like an illiterate person, because I can't write by hand. I can read stuff I can't write. Even hiragana. I never did enough practice. My wife writes down everything so easily.. and I can't. Writing romaji.. argh. I hate that I can't write, by hand, what I can read.
- latentsea 2y agoI think hating that is kind of a choice. Like sure it's bothersome, but if you didn't grow up there and spend a decade in school doing a lot of writing by hand then the only substitute is to do a crap tonne of writing by hand as an adult. The investment in time for such a small pay off as filing out forms being less of a hassle likely isn't worth the full investment of time required. So, if you just accept it for what it is, then don't mind the extra time required when filing out forms, I guess it's not so bad.
- kevin_p 2y agoIt's designed for Japanese where it probably works better - pretty much all Kanji have multiple pronunciations that can be completely different to each other so it makes some amount of sense to ignore the sound and focus on the shape and the meaning. Much less relevant for Chinese where you can usually tie the character to a single sound and learn all three halves at the same time.
- cyberax 2y agoIt surely helps. Mnemonics are a good tool for memorization. You don't have to remember elaborate stories, just something like: "歌" - "older brother lacks singing". It certainly doesn't work with all characters, but it helps.