Blog Archive

Tuesday, December 16, 2014

When small changes make a big difference, part 3

This is the last and final post about confusingly similar Chinese characters. In the first article, there was a quiz where you could check how many of these characters you know, so if you haven't taken the quiz yet, I suggest you do that before you continue reading this post.

Last week, I explained the answers to the first half of the confusingly similar character pairs and triplets, this week it's time for the remaining seven characters.

土 (tǔ) "earth, soil", 士 (shì) "warrior, scholar"

As was the case with 未 and 末, the only difference between these two characters is the length of the horizontal strokes. The first character, 土 (tǔ) "earth, soil", has a short stroke on top and a longer stroke at the bottom. It's a common character (624th), both used on its own and in compounds such as 土地 (tǔdì) "land, territory" and 土豆 (tǔdòu) "(Mainland) potato, (Taiwan) peanut". It's also one of the five elements.

The second character is 士 (shì) "warrior, scholar" and has the opposite strokes, i.e. a long one at the top and a short one at the bottom. This character is also common (368th) and means "warrior“ in some common words, such as 士兵 (shìbīng) "soldier", but is more related to studying in other words, such as 博士 (bóshì) "doctor, Ph.D.".

Since these characters mean completely different things, there are no real communication issues if you get them wrong (the reader will be able to guess what you meant), but if you want to write correct characters, do pay attention to the length of the strokes. This of course includes when these characters are found as component parts in other characters!

 夭 (yāo) "young", 天 (tiān) "sky"

This example is similar to 千 and 干 in that the difference is the slope and direction of the first stroke. In 夭 (yāo) "young", the first stroke should be written from right to left and slope gently down, whereas in 天 (tiān) "sky" the first stroke is a normal horizontal stroke from left to right.

夭 (yāo) "young" is not a common character in itself, but it is contained in some common characters, such as 笑 (xiào) "laugh, smile". Thus, if you write the bottom part of that character like you write 天 (tiān) "sky", you're not doing it right.

天 (tiān) "sky" is of course a very common character (55th) and appears early in many textbooks. It can also mean "day" and is most commonly seen in words like 今天 (jīntiān) "today", 明天 (míngtiān) "tomorrow" and 昨天 (zuótiān) "yesterday".
入 (rù) "enter", 八 (bā) "eight", 人 (rén) "person"

The last group of confusingly similar characters is perhaps the most confusing one. All three characters are common both as individual characters and as character components, making it hard to remember which one is which. Knowing their basic form and meaning makes it a lot easier to remember which of them should go in a specific compound! Rather than trying to remember details of the strokes, it's better to remember the meaning of that component.

入 (rù) "enter" is the 193rd most common character and is written first with a down-left stroke and then the down-right stroke, obviously longer and leaning over the first stroke. It also has a slight hook at the beginning. This character is most commonly seen in words like 加入 (jiārù) "to join" and 进入 (jìnrù) "to enter".

八 (bā) "eight" should be familiar to anyone who has studied for more than a few weeks and is the 282nd most common character. It is written with the same stroke order as 入 (rù) "enter", so the only difference is that the two strokes shouldn't touch each other. Also, the second stroke doesn't have the characteristic hook that make 入  easy to recognise. Sometimes, "eight" is also written with more vertical strokes of almost equal height There is also a fraud-proof version of 八 which is harder to confuse (and change): 捌.

Finally, 人 (rén) "person" is also a very common character (6th) and it's usually one of the first characters students learn. Note the difference between this and the other two characters, though! 人 (rén) "person" is written with the same stroke order as the others, but the first stroke should be longer and leaning over the second stroke that supports it. In a way, it's the opposite of 入. Apart from being a very common character in words, 人 is also common as a character component, but pay attention, because it might change shape!

What characters do you find confusingly similar?

I have now covered seven groups of confusingly similar characters and I hope this will help you write better characters and confuse similar ones less. There are more similar characters out there, though. Do you have a set of two, three or even four characters that you find annoyingly similar? Leave a comment and share!

Thursday, December 11, 2014

When small changes make a big difference, part 2

Last week, I wrote a post about how small changes could make a big difference in the meaning of some Chinese characters. I also wrote about how difficult it is as a beginner to figure out exactly which stroke lengths, placements and angles are crucial for determining meaning and not just if the character looks good or not.

In that article, I also included a quiz with fourteen characters that are easy to confuse with one or two other characters that differ only slightly. Here are the characters I used, but I encourage you to go to the previous article and do the quiz before you continue reading this one!

己/已/巳, 未/末, 千/干
土/士, 夭/天, 入/八人

Below, I'm going to explain the differences between the characters in the first three groups, the rest will be covered in a similar article next week:

己 (jǐ) "self", 已 (yǐ) "already",  巳 (sì) "the sixth heavenly branch"

The difference between these three characters is where the third stroke starts

In the first case, 己 (jǐ) "self", it shouldn't go beyond the horizontal second stroke (it still does sometimes, but it's just the beginning of the stroke that protrudes above the second stroke). 己 is a common character (131st) and appears most commonly in the word 自己 (zìjǐ) "self, oneself". This character is also common as a component in other characters, such as 记 (jì) "register, record".

In the second case, 已 (yǐ) "already", the third stroke clearly starts above the end of the second stroke, but it doesn't touch the first stroke. This is also a very common character (95th) and most commonly appears in the word 已经 (yǐjīng) "already". Think of it as the stroke as "already" gone past the horizontal stroke.

In the third case, 巳 (sì) "the sixth heavenly branch", the third stroke goes all the way up and joins with the first stroke. This character is not so common on its own (it's not even among the most common 3000 characters). However, it is common as a component in other characters, such as 包 (bāo) "wrap".
 末 (mò) "tip, end", 未 (wèi) "not yet, have not"

Both these characters are based on the character 木, which is a pictograph of a tree. By adding a vertical line marking the top of the tree, the writer could represent the tip or end of the tree (or of something else by extension). This character is fairly common (1272nd) and is taught in most beginner courses in the word 周末 (zhōumò) "weekend", i.e. the end of the week.

The second character, 未 (wèi), is much more common (399th), especially in written Chinese. It means "not yet" or "have not" and the first example students learn is typically 未来 (wèilái) "future", i.e. something that has not yer arrived. This character is very common as a negation in written Chinese and there are many, many uses of it, but here are two examples: 未知 (wèizhī) "unknown", 未必 (wèibì) "not necessarily".

The only difference between the two characters is the length of the horizontal strokes. I remember these two by thinking of the origin of 末; the longest stroke marks the important part (i.e. the tip of the tree).

千 (qiān, thousand), 干 (gān/gàn)

These two characters are very similar and can be easy to confuse, especially when reading. Writing is not so bad, because they are written with different strokes.

The first character, 千, is written with the first stroke sloping down and left, whereas the second character 干 has a normal, horizontal, left-right stroke. This is a good example when stroke order matters! If you use handwriting input, you're much more likely to get the right character if you get the stroke direction right. In other words, the characters are visually similar, but written differently.

千 (qiān), "thousand" is a common character (410th) and usually appears in numbers. There is  also a fraud-proof variant of this character, which has an added person radical: 仟.

干 is more complicated, because it's the simplification of several different traditional characters (four, actually, 乾, 幹, 干, 榦, but the last two aren't very common).
  • 乾 "gān" means "dry" and is first encountered in the word 乾淨/干净 (gānjìng) "clean".
  • 幹 "gàn" means "do" (and also means the f-word) and is most commonly heard in phrases like 幹嘛/干嘛 (gànma), meaning "what are you doing".
 Thus, in simplified Chinese, the character 干 has several completely different usages with difference pronunciations! The confusion between 干 and 千 seldom occurs in traditional Chinese because 干 isn't very common.

The remaining three groups will be the subject of a similar article next week. By way of rounding this one off, though, I'd like to address a key question. Does it matter? If you write 己经, 周未  or 两干块, most people will understand what you mean, they might not even notice the incorrect characters if they don't pay attention. So why bother?

There are several reasons. First, some of the differences are crucial when dealing with handwriting input. If the strokes are in different directions, such as for 千 and 干, the input method won't recognise the character if you get it wrong (for instance, you can't write these characters using Google's handwriting recognition with the wrong stroke direction).

Second, writing things correctly gives a neat and thorough impression. You can certainly understand what I want to say in English if I write "They have recieved teh pakcage", but I think most of us share the idea that following a common standard is good and generally improves communication.

That said, the only place someone is going to really care what you write is probably in a classroom or when you sit an exam. This doesn't mean that it's unimportant, but it means that you need to decide for yourself how thorough you want to be!

Friday, December 5, 2014

When small changes make a big difference, part 1

Some strokes that make up Chinese characters are very important, drawing them just a bit too long or with the wrong angle will change the meaning of a character completely. Other strokes are not so sensitive and writing them incorrectly will just make the resulting character ugly.

Below, I have created a quiz with 14 similarly looking characters that differ only in the length or slope of one stroke. See how many you know! If you have any suggestions of more pairs or triplets, please leave a comment!

Part 1:

1. 已
a. yǐ, already
b. sì, the sixth heavenly branch
c. jǐ, self

2. 己
a. yǐ, already
b. sì, the sixth heavenly branch
c. jǐ, self

3. 巳
a. yǐ, already
b. sì, the sixth heavenly branch
c. jǐ, self

Part 2:

4. 未
a. mò, tip; end
b. běn, root
c. wèi, not
5. 末
a. mò, tip; end
b. běn, root
c. wèi, not

Part 3:

6. 千
a. yú, at; in
b. qiān, thousand
c. gān, dry

7. 干
a. yú, at; in
b. qiān, thousand
c. gān, dry

Part 4:

8. 土
a. tǔ, earth
b. shì, scholar; warrior
c. gōng, work

9. 士
a. tǔ, earth
b. shì, scholar; warrior
c. gōng, work

Part 5:

10. 夭
a. yáo, young
b. tiān, sky
c. tài, excessive

11. 天
a. yáo, young
b. tiān, sky
c. tài, excessive

Part 6:

12. 入
a. rù, enter
b. rén, person
c. bā, eight

13. 八
a. rù, enter
b. rén, person
c. bā, eight

14. 人
a. rù, enter
b. rén, person
c. bā, eight

Here are the answers, but do make an effort before you peek!

 Key: 1a, 2c, 3b, 4c, 5a, 6b, 7c, 8a, 9b, 10a, 11b, 12a, 13c, 14b

How did it go? If you're a beginner, don't be too discouraged by this, these really are some of the trickiest cases. If you're an intermediate learner, you should know at least half. Advanced learners should get all or almost all of them!

 When small changes make a big difference

Learning to separate characters like these requires what I call horizontal vocabulary learning, which means that you can't just drill down (look at character components) or go higher up in the hierarchy (looking at what characters and words a character appears in), you have to look at similar characters at the same level as well. This requires focused studying, even if it can of course be accomplished by a very large amount of exposure as well.

Here are two follow-up articles to this one, covering the answers to the quiz in detail:
Before I round this article off, though, I want to share with you a game I usually play with the students when I teach beginner courses in Chinese.

Chinese whispers - with characters

The idea of this game is to show the students how quickly they go from viewing Chinese characters as pretty pictures to regarding them as symbols with a structure that is part of a larger writing system. The game is based on a well-known game, which is rather suitably called Chinese whispers.

The original idea is that one person starts with phrase, then whispers it to the next person, who then whispers it to the third and so on until the message reaches the last person, who says the words aloud. The message is usually hilariously corrupted by this time, which is the point of the game.

In this Chinese character variant, the first student doesn't whisper words, he or she writes characters on paper and sends them to the next, who then copies the character on a new piece of paper and sends the copy on. After passing ten students or so, most of the original characters are completely unreadable for the teacher. The reason is simple. They don't know what matters. They think they're copying accurately, but they focus on the wrong things. They think a certain feature is important, but in fact it isn't. They underestimate the spacing of the components and the length of strokes.

The interesting thing is that when I play the same game with them a week later, after teaching the basics of Chinese characters, all teams can easily copy the characters ten times without corrupting them beyond recognition. It also takes about half the time it takes the first time around.

Thursday, November 27, 2014

Tending your vocabulary garden

Image source:

If I could go back in time and change the way I learnt languages in general and Chinese in particular, there are many things I would do differently. In fact, a lot of the things I write about language learning these days are based on what I remember myself doing or have observed other people doing while trying to learn a language, then relating that to what I know about the subject today.

One of the areas I think I did worst in when I started learning Chinese was vocabulary. I don't mean that I didn't learn enough words or that I used a horribly inefficient method, I think I did okay. The problem was that I learnt the wrong words and I was stubborn enough not to fix that for years. This is what I want to discuss in this post.

Your vocabulary is like a garden - it needs tending

I think it helps viewing your vocabulary as a living thing; a garden seems like the best analogy. Plants in a garden take time to grow and the overall results benefit from planning ahead. A garden can also have different functions and will require different plans depending on what that function is.

Most importantly, you need to keep an active relationship with the plants in your garden, identifying which plants to keep, which to prune and which to get rid of altogether. If you let everything grow without control, your garden will turn into wilderness. Vocabulary should be regarded in this way, too.

Some common problems and how to avoid them

The most common mistake I see with ambitious students (including myself years ago) is that they tend to think only in terms of quantity: the more the merrier.  While it might be true that knowing many words is good in general, this isn't always the case, especially not in the short run. At least, it's not that simple.

The problem is that you only have limited time available to learn and maintain vocabulary, so it matters greatly which words you choose to learn. Of course, it's hard to know which words are important before you know them. A native speaker or an advanced second language learner can tell you which words are essential to know and which aren't, and sometimes frequency data can give you useful clues, but on your own, knowing which words are worth adding remains a problem.

The best way of dealing with this problem is either to add words from trusted sources or to prune and manage your deck actively. Doing both is also an option.

Adding vocabulary from trusted sources

The first solution involves mostly reading material that has been designed for you or someone at your proficiency level. This includes textbook, graded readers or other learning materials aimed at second language learners. In these texts, you're unlikely to find extremely rare characters or words, so you can be relatively confident that what you're learning is useful.

Image source: http://atomiclemon.
Tending your vocabulary garden

The second solution, actively managing your vocabulary, is a must as soon as you start approaching authentic texts. They will contain a very large number of words you haven't seen before and adding everything to Skritter isn't going to work. Even if you spend the time necessary to accomplish that, it still wouldn't be a good way of learning because you would waste time learning words that aren't actually improving your overall language proficiency that much.

Actively tending your vocabulary garden is important. Delete words you don't like, that seem less important than when you added them or you think are slowing you down in general. Edit any character or word that you don't like or that isn't clear enough. Save other types of information you pick up about characters and words (the easiest way to save these is by editing the custom definition).

Ready-made gardens and vocabulary lists

This is why I don't really like importing lists created by other people. If I wanted a garden that suits my preferences and needs, I wouldn't go online and just download a ready-made one. Similarly, unless you're a true beginner, other people's lists comes with problems attached.

I do sometimes use lists created by others, but mainly to find a certain kind of vocabulary (based on a textbook for instance). In these cases, I always spend a lot of time making the list my own. I edit the definitions, delete things I don't need and add things I think are missing. I don't just add the list and expect it to merge with the vocabulary I already know. Naturally, editing a list takes time and I don't do everything at once; everything is done on a need-to basis.

Growing your vocabulary

Following the advice in this article would have saved me a lot of time and accelerated my learning quite a bit, all the way from the beginner level up to advanced. It's a grave mistake to think you have to catch them all or try to learn every single new character/word in a text you're reading. Your time is limited, don't waste it learning words you don't really need!

Thursday, November 20, 2014

Kana is here! Now available for the Android app.

Attention Japanese Skritterers: Over 5000 lines of code and countless coffees later, we are excited to announce kana handwriting support is finally available for our Android app. We said we won't stop making Skritter better until Chinese and Japanese are easier than French, and we mean it!

For those reading who don't know, Japanese uses three writing systems, hiragana, katakana, and kanji. We've written a blog post about the their history and usages if you want to know more.

Chinese users have long been able to use Skritter to practice every word that they study fully, and now for the first time ever, Japanese users can too. Before kana handwriting support was available for the Android app, Japanese users were forced to skip the writing of any kana, and instead only write the kanji. Now with kana enabled, you can write words in their entirety. You can even write out full grammatical sentences if you want to!

This opens up a whole new window for Japanese students. Before, a student on Skritter would have to be already familiar with kana in order to read the pronunciations for words, (romaji is a hindrance for learning, which is why we never supported it). Now it's finally possible to study and write kana. You can start from scratch without knowing how to read or write a thing. This is a very important improvement for Japanese learners!

The feature is not yet available for iOS or the web, however we are eager to implement this if it proves popular! 

If you are brand new to Japanese and haven't yet learned kana, you can start studying from these lists (it's recommended to learn hiragana first):

Skritter | Hiragana
Skritter | Katakana

For now, the option is disabled by default, so if you want to give kana handwriting a go, first enable the setting from within the Settings area of the app:

Since the feature is so new, there may be a few kinks still to work out. If you have any comments or suggestions for improvement, it would be greatly appreciated if you could let us know.

Wednesday, November 12, 2014

Study Smart, Not Hard

Most who start studying foreign languages become infatuated with learning new words and expressions, and seeing the differences and similarities compared to their native language. Unfortunately after time though, sometimes these exciting feelings begin to dwindle, and in some cases they might even lose all interest because it doesn't seem as new or interesting as it first might have. They might claim to have the feeling of "not learning any more" or "not having the time" to maintain what they've already learned.

The feeling of "not learning anymore" reminds me of the comparison of shaving a head bald. Starting off with zero hair whatsoever, then going to bed and waking up the next morning with a head full of stubble seems like a drastic difference, (the largest difference there actually will be in the hair regrowing process).

It's sort of like starting off with no words in a language (a bald head), and then learning your first 10 words (gaining some stubble). You then learn your next 10 words, doubling your vocabulary which seems like another big change, just like how the second day of hair growth doubles the length of your total hair. Eventually you get to the point where you have so much vocabulary, or so much hair, more doesn't seem to add any length.

However the rate is still consistent of course, it's just that compared to 1000 words learned, adding 10 more words doesn't seem like a such an amazing feat anymore, or when having 2 feet of hair, adding another inch isn't even noticeable and it seems like your hair stopped growing.

Some "study habits" I like to keep have kept Japanese just as exciting, and I'm sure can be applied to any language. The key is you shouldn't really be "studying", instead activities you like to do anyway can be in Japanese. If you're not ready for that step, then it should be increasingly Japanese, replacing what you can and learning the vocabulary.

If you prefer your activities in English and think it's overboard to replace your activities in Japanese where possible, then it's a good chance you don't want to be learning Japanese anyway! (Or at least get to a proficient level). It's a thinking platform for the brain after all, it can be used at any point in time, any point in the day... for anything!

Here are a few tips I've learned:
  • Do any reviews you would normally do in the day first thing in the morning, and very last thing at night-- even if it's only 5-10 minutes each. Keeping it fresh in your mind first thing and getting it out of the way is a good tactic. It also seems the mind is most creative in the morning and fades as you start to fully wake up. A lot of musicians believe in this theory and practice right as soon as they wake up. It's recommended to do this as soon as possible right after waking up, as in, stay in bed and don't get out until you finish. 
We all have obligations that interfere with things we actually want to be doing, like learning Japanese if you're required to use English throughout your day-- so bust out some reviews ( in the morning and what you studied will stay with you throughout the day. That way when you go back to review and study more at night, you're solidifying any information remembered from that morning.

It's an easy to form a habit which really boosts retention. I don't recommend pushing it with the reviews though, reviewing words in a spaced repetition system is very important, but after using the same "study method" hours on end day after day, you might not want to go back the next, which defeats spaced repetition systems' efficiency. It's best to keep it short and sweet everyday, opposed to doing a long day and then missing a few days, even if you are averaging the same overall amount of time.
  • Don't study component characters-- a component character on Skritter is kanji that contains every reading and meaning for that specific character. Studying words gives you context of when that kanji is actually used and how it is pronounced in the word. You still can learn the individual meaning through the detailed information screen or in the context of the actual word. It's great to learn kanji and know every associated reading and meaning like an 生き字引き, but it's also really great to actually know Japanese words, and which characters are frequently used.
  • Avoid studying! Watch shows, listen to music, read a news article, book, or manga, anything interesting to you, and pick out words you don't know and write them down (or even better quick add them to a list on Skritter, which both looks up the definition and saves it to a list to study). Think of it as being a little kid again, and rediscovering all of the things you had once before with your native language. I imagine a lot of kids read comic books and may not understand all the expressions and words, but it's through reading or listening anyway and doing fun things that you learn, not by cramming a bunch of words out of context and sitting at a desk. I would add them to Skritter or where ever and study them, but also be sure they are coming from fun sources. Chances are you are going to hear those words again and quite frequently, so why not learn them now? 
  • Try not to rely on furigana. If you are reading something give yourself a chance before hovering over the word with Rikaisama (or whatever popup dictionary you use) . You might surprise yourself! Try to guess the pronunciation and meaning before you check, you might be able to decipher it if you keep thinking before immediately grabbing the answer, and it's a good feeling to figure out a word you didn't know by context alone.
  • Don't use subtitles. Never use English subtitles, and if you must, use Japanese subtitles instead, but eventually you don't want to use Japanese subtitles either. If you are reading subtitles in English but "listening" in Japanese, (which doesn't really work), you may indeed catch more words in Japanese that you wouldn't normally hear without any, since there's more catchable context, it may make you feel like you're reenforcing memory or understanding more, however it's actually a very bad habit which should be avoided. 
It's along the lines of flash card or dictionary syndrome, where someone can gain the ability to memorize 20,000 words if presented with flash cards, but phenomenally won't recall them when presented in the wild. The only exception would be if you aren't yet at the level to enjoy watching without English subtitles enough to continue watching, in which case watch with them on if you must opposed to not watching at all, but don't use it as an excuse to not work towards the level where you can start watching without!

I recommend trying to get to the point where you can switch to Japanese subtitles and look up all the words you don't know. Appropriate difficulty level is important, if you are watching a bunch of investigator shows you might learn technical words you aren't "ready" for, so I would go for movies and shows that don't seem like they would use a bunch of advanced vocabulary. After you watch through them with subtitles, you'll likely remember much of the plot and what people said, so try to give it a go without any subtitles. Repetition is unfortunately the key to mastery with things, as annoying as it may be it's extremely effective.

Once you get to the point where you can regularly understand most of everything being read in Japanese subtitles, it would be time to remove the subtitles and resort to backing up the video every time you don't understand something, and if you still can't grasp it to look at the spot in the Japanese subtitles.
  • Ban leeches-- if you constantly keep getting a word wrong but it keeps popping up, just ban the word and then unban it after time passes. Sometimes it seems like there is some self sabotage on any chance to remember specific words or writings, which usually passes with time.
  • Learn sentences and memorize them, of course making sure they are from reliable sources. Try to memorize 20 sentences a day ideally and also at max, using mostly words and grammar points you already know but adding a few new things here and there. It's especially helpful to write them out by hand. It's great if you have a study partner you could practice with verbally, where you can quiz each other by either reading the sentence in Japanese and quiz for the English, or recall the Japanese by hearing the English first. 
Studying with a partner of course makes it feel less like studying. I said avoid studying, but you still have to do it of course! As a note, if you do use English sentences first trying to recall the Japanese equivalent, it can be a bit annoying to memorize the exact Japanese sentence, especially when you get to the point where you naturally learn to state things in different ways. (That becomes more like memorizing a line from a script, and less like learning Japanese).

The exception would be if your study partner "grading" you was a Japanese native also fluent in English that could verify your version was the same meaning and nuance as the English definition. This is why hearing the Japanese, saying what it means in English, then repeating the Japanese is recommended as a good start. If you are fortunate enough to study with someone fluent in Japanese and English, it's much more effective to do English to Japanese, while trying not to "memorize" a sentence (if that makes sense!).
  • Read like crazy-- you should read everyday, but at the appropriate difficulty level. If it's too difficult to read at your Japanese level, you could start off by memorizing stories, breaking apart the words and grammar sentence by sentence, word by word, starting from the very beginning, and then re-read them everyday until you're absolutely sick of them, then advance farther along in the story. It's very helpful in the beginning stages to reread material over and over again until you understand it, which usually comes with the unfortunate ability to blindly recite it. Once you're fairly proficient you may not want to read over the same thing again since it can be a bit tedious, and instead pick something that isn't too difficult to read and progress through it. 
There are a lot of fairy tales which aren't too hard, (though may have some expressions in them that aren't typically used outside of writing) over at: If you go to the world stories area (世界の昔話), you can find stories you're likely already familiar with, which can be helpful for context so you aren't completely lost if it's difficult to read or comprehend.

If you don't have a popup dictionary allowing you to hover over words on the web giving you the definition, you should give it a go and install one. I personally think that Rikaisama for Firefox is the best bet, but there are other versions like Rikaichan and Rikaikun as well.

Another great tip is to use Skritter's bookmarklet tool which lets you highlight a word on the web and add it to a lists, which you should do for words you don't know. You can add the bookmarlet via the right hand lower corner of the home page at:
  • Have conversations! If you don't know anyone that is native (because it's backwards progress to have conversations in broken Japanese),  try meeting people from websites like Lang-8 or Japan-Guide. You will find so many people wanting to learn your native language and willing to help you learn theirs. The idea on Lang-8 is, you write something in the language you are learning and native speakers of that language will either correct your mistakes, or even better, give you another way to phrase what you are trying to say to better match the nuance of what you meant, or to sound more native. A lot of the time what others write is in their native language in addition to the language they're learning, which gives the chance to learn by reading something a native wrote, and at the same time sharing a good deed by correcting theirs. People seem to be super friendly on Lang-8 and it's fun to read what others write about. Make sure you actually read and correct others work though, not only is common courtesy but they will likely return the favor! A lot of Lang-8 users are interested in video chats on Skype if you're looking to practice real time conversation.
Just try to remember to have fun learning the language you want to and you'll keep progressing. If you have any good tips please be sure to mention them in the comments!

Friday, October 31, 2014

What should you do when you forget a word?

Image source:
A spaced repetition program such as Skritter is built around efficiency. The goal isn't to make you remember everything, it's to make sure that each minute you invest will give you as big a return as possible. In the long run, this will improve your learning a lot.

In Skritter, you can adjust the retention rate, i.e. how large a percentage of the vocabulary you want to remember. You can do this by going to your account and changing the study settings. This roughly means that if you use Skritter regularly with the retention rate set to 92%, you should know roughly 92% of the words at any given time.

This sounds bad to some people. If you have 1000 words and you forget 8% of them, there will be 80 words you don't know! Relax, that's actually good for you. If you could set the target retention rate at 100%, you would waste a lot of time reviewing words you don't really need to review. The closer to 100% you come, the more time you need to increase each percentage point. Still, since remembering words is the main goal with using Skritter, you can't lower the settings too much either.

What do you do when you forget a character or word?

The upshot of the above discussion is that a healthy study strategy includes forgetting a relatively large number of words. However, I think it's a mistake to just treat this as part of the natural process of learning and just continue studying. It does matter what you do with the words you forget.

Let's look at a few common scenarios:
  • Pretend that you actually knew the word and grade yourself better than you deserve. That's a form of cheating where there is no winner and you are the only loser. I've written more about this here. You should never do this, it's terribly bad for your learning in the long run, even though it might feel better in the short run.
  • Mark the word as forgotten and keep reviewing, leaving the failure behind as quickly as possible. In the worst case, you don't even look closely at the word you failed to see what and why you failed, you just click next immediately and trust you will learn the word later. You might, but it's also likely you keep doing this for a number of words in your deck, wasting time every time you do so.
  • Look at the word and actively determine what should be done with it and take appropriate action. This is what I think everybody should be doing, but I have a feeling that it's not as common as it should be. This requires some elaboration. Having an active attitude to vocabulary learning is very important.
Actively dealing with forgotten characters and words

The most important word in this subheading is "actively". This means that you can't treat all characters and words equally, you need to adjust to each situation and base your action on many different factors.

Here are a few things you can do with a forgotten character or word:
  • Delete it. Do you need this word? If you probably don't, delete it. If you're not sure, delete it (it will surely appear again in your listening and reading, and you can add it again later if you want to). If you know you really need the word, keep reading this list.
  • Edit and upgrade the word. Actively look at the information you have and make it better. Create a new mnemonic (the last one obviously didn't work), add a custom definition or expand the current definition. Draw a picture. Spend some quality time with the word.
  • Drill down and learn more. Look up the individual characters in a word or the character components in a single character. Use another dictionary if the current definition doesn't suit you. If you're confusing two characters or words, look them up, you're unlikely to be the first to confuse them.
  • Add supporting characters and words. If you can't get a character to stick and it's an important one, add other words that also contain the character. If you keep forgetting how to write , add several common words containing that character. You will see it a lot in the near future and you have a chance to really get to know it (and the new words, of course). You can add words from Skritter (web version) or use Hanzicraft to get a quick list of frequency-sorted vocabulary (for that would mean 凝固, 凝望, 凝聚 and 凝视 (凝視).
I could make this list longer, but I think I've made my point. The decision what to do with a forgotten word usually just takes a few seconds, but will lead to a much-improved retention rate without spending too much extra time. Actively trimming and adjusting your collection of words and characters also makes sure that you don't waste time on vocabulary you don't actually need.