Memorizing characters in pinyin structured table as 'investment' in learning Chinese better

[Pall]

Why characters first?  Chinese words are not words in common sense. They are rather "stable combinations of characters", than words. And some of them are not as stable as others. I read that in a linguistic study and tested it on my own experience. 

 

All Chinese words consisting of 2-4 characters arose from more simple words of only a single character, and the meaning of the words while their creation was associated with the meaning of the initial characters historically, though today in many cases it's not easy to derive it from them. 

 

Thus, there are objective reasons to split the words into charcters and learn characters in addition to learning words or even before that.

 

Also there are strong practical considerations to do that. When one learns new words in alphabetic languages he does that already knowing the letters. But in Chinese if one encounters a word containing new characters for him, he has to do quite different tasks simultaneously, to learn the characters and to learn the words in their meaning. It's one of the main reasons why it's so difficult to learn Chinese. But it can be easier if he learns "the alphabet" first. 

 

When we need to put in order a long list of different objects in our mind we should divide them into groups firstly by the feature, which varies less. For example,  objects vary in shape, color and size, and there are 5 sizez, 20 colors and  100 shapes. It's logical to start with the size, then classify the objects by colors and finally by shapes. If one starts with shapes he will embark on the hardest road. That's what some do when they begin with the meaning of the characters, or their structure, postponing the task to learn the sound or even rejecting it at all. 

 

There are only about 400 pinyin syllables without accounting for tones and 1200 syllables with tones, and only a part of them is used for more common characters.

 

Please, find below a table of the HSK5 characters grouped according to their pinyin transcription. It helps to memorize characters, as well as words eventually, better, though some of the potential is linked to the knowledge of Russian. I think, even the first Excel page of it is rather helpful, and one can find his own ways to learn those characters.

 

 

Of course, Chinese "alphabet", if to take characters in this way, is too long. But why not lo learn firstly those letters, which are used in more common words? I came to a conclusion that a good measure of the "common" would be the HSK5 list (HSK4 looks as starter's, whreas HSK6 is too advanced even for many Chinese, as I can guess).  And when you know that, for example, guān can only be 观, 关 or 官 in the HSK5, it's morally easier to remember these characters when you meet them for the first time in your texts. And when you know that guài is present in the list only in the 怪 form, you can be just happy. There are, certainly, other cases, with 16 different characters for shì and 14 for jì. But on the everage it's only 2.03 charcters per pinyin syllable.  The first page of the table, which I uploded, looks as in the picture below. In the uploded version all characters are in black, though.

 

I marked in violet the known characters and in red those characters that I'd like to learn first.  I chose such taking into account their simplicity and how many components well-known for me they contain. Further, I made another page that include only the known characters and those, which I'm going to learn first. The picture is shown, too. Now I can observe the whole range of my knowledge and would-be knowledge, and to notice some regularities. Please, note, that some characters are shown several times according to their different pronunciation, these are marked with borders. Also, I included only those variants of the neutral tone, which represent seperate characters or cases when the character is used as a syllable of words only in neutral tone (at least in the HSK5). 

 

I've found, that learning characters in groups according to their pinyin serves as a mighty method to remember them. In the most difficult cases one  may apply functional analysis to remember the characters relating to the same pinyin syllables. But I'm sure, it's not necessary. Instead of that I made third and forth pages.

 

The third page contains  Russian prefixies reflicting all the components of the characters, showing the remaining small details iwith the word "экстра" (extra) in the end when necessary. The screenshot is below. Now I can, reading the prefixes, to imagine the characters in my mind. It's worth to say, that in Russian these prefexes look quite natural, and they're not too long if considering for the amount of the information they include. Only word roots are used here with "o" or "e"  links. In Russian there are many words, consisting of two roots, like паровоз - wapor+to carry, a locomotive, very sinilar to Chinese word creation. There're some such words in English, too. 

 

In the fourth page one of the characters corresponding to a syllable is regarded as the main one, it stands on the first place (one that I know better). In the colomn right to the pinyin I give its meaning in capitals. There can be several meanings, but I chose only one, not similar to the meaning of other characters in order not to confuse. For example, in the word
帮助 both characters mean "help" or "to help", so we should give this meaning to only the first one, and for the other to use its additional meaning as "assistant". Other characters in the line corresponding to a syllable are also shown in one of their meanings. Then I created a formula, a phrase that would link the meaning of the "ordinary" characters to the meaning of the main one. For example, for 不/部/布 it is something like "Not a ministry of fabrics". This phrase is written as a note for the first translation, the option being provided be Excel in its latest version ( there is an oppostunity to write a remark and also a general note, I chose the latter form). In the note I put the main word in the first place (in Russian any word order is possible) and underlined it. The words corresponding to the other characters I showed in Italic. 

 

Also there is a coloumn to the left from the pinyin. It's for two things. First, it includes some general idea, linking the characters in the line with the main one. Second, the first letter in the pharse shows the number of characters in the line (initially I showed that with the first letter of last word). It's important to feel the range even when you meet new characters, not present in the list. You may add them to it, but it's better to mark them as "extra". 

 

You may find that the list is not up-to-date. They often change the lists of HSK words, transfer some words from 5 to 6 and back. But it doesn't matter, the uploaded list is just for orientation. You should take somethig as the basis, and that's all. 

 

As to words, consisting of 2-4 characters, we can always connect their meaning to even the single meaning we chose for the characters. And when you have learnt  characters, it's always possible to give additional  meaning to them. 

 

I hope it's helpfull.

 

The best use of the list can be in combination with the procedure to work with texts described by me in https://www.chinese-forums.com/forums/topic/58727-characters-learning-how-to-proceed-after-radicals/page/3/#comments. Without knowing Russian it also can be used partially.

 

 

 

 

HSK5 syllables (pinyin&characters).xlsx

[Pall]

In the table the pinyin syllables starting with "ch",  "sh" and "zh" are given separately. Thus, there are 26 letters or their combinations with which syllables can begin.

[Pall]

According to research results, human brain can remember up to 7 simple objects, not connected with one another, in a row (numbers in telephone numbers, etc.). Therefore for shì  and jì it's better to designe two or three furmulas for each.

[Pall]

And at the final stage of learning there shoud be typing (and later hand-writing) the Formula with changing Russian words for the character meaning back for characters. "Собирать украшения, наступая на них (To collect jewelry while stepping on it)" = "采彩，踩 на них". It's like a superscript, about which I wrote earlier, but this time according to Russian grammar, not Chinese. In Chinese the sequence of these characters doesn't make sence, of course.

 

Here are other examples of final Furmulas with characters.

 

Comprising three characters:

В нашей 班 般 людей 搬ся.
比彼 чужой 笔 со своим.

不_部_布.

差 между 插叉 или нож в зад нулевая.

 

Comprising five characters:
长 рука 常，场 снегопады, 尝 и 偿  долги

 

What is importnat, it's not just mnenonics. The characters are linked in their MEANING in the context, though created in another language. And such connection will not be necessary to remember these characters very soon, you'll be able to name characters with the same pinyin similar as you can, for instance, recollect in your mind all men named John, whom you know. 

Edited September 24, 2019 at 06:01 AM by Pall
additional information

[Pall]

 It's reasonable to learn firstly charcaters for those syllables, which  five and more characters correspond to, I've counted 58 such pinyin syllables. It's easy to remember them since the pronunciation is the same, on one hand.  Less work brings more fruit. The number of characters corresponding to them is close to one third of the total. And on the other, later it'll be possible to compare other characters to them and make intersecting associations. 

[Pall]

How to work with the formula better. 
First, one should read the formula, recollect the characters and type them. For example:

Разговор такой: под дождем предоставлять крышу, особенно неоперившимся - The word is: when it's raining it's necessary to provide roof, in particular to eyas.
One is typing:  语   雨   与   宇   羽.

Let's assume, he has difficulties with 宇 and 与, while knows very well 语 and   雨. Then he should rewrite tha phrase, combining his language with Chinese and putting less known characters close to the well-known. 

语 такой: 宇 под 雨 надо 与 особенно не羽. 

 

In general, one should type these rows of characters several times in different order. It is necessary to achieve the association of equally sounding hieroglyphs with each other. Besides, this will be important  when it's necessary to distinguish them in someone else's speech.

[Pall]

Well, anyhow or other I've learnt 80 new characters in two days. (THere were 90 characters in the set, but I knew 10 of them before and used them as the main characters to which I linked the knowledge of the others for the beginning.) I can not only recognize them, pronounce and recollect their meaning, I also can say how many characters in the HSK5 correspond to the same pinyin syllable, and even type all the other characters having the same transcription within the HSK5 list, when I see one of them. I started from the end of the table,  'zh' , and covered 'y' as well. At this stage I 'm learning only characters for those syllables, to which five and more characters correspond. They were zhù, zhì, zhǐ, zhí, zhī, yuè, yuán, yù, yǔ, yú, yóu, yì, yè, and yán, 14 syllables. For  'z' ( 'zh' is separate) there were no such ones. 

 

It seems, I'll be able to learn the whole table in two-three months. I want to know it like chemists know Mendeleev's periodical table. I expect, that it would give me a great freedom to remember new words, grammar then, for the basic boring work would be done already. Of course, there will be other new characters, but I'd have a solid basis in my mind, and they would be just annexes to it.

Edited September 24, 2019 at 03:55 AM by Pall
additional information

[Pall]

In addition for better visual memorization all characters with the same pinyin within the list can be distributed in space like this example for 10 characters sounding yù. It also helps to remember the number of the characters. But it's better to do later when the characters are memorized by other means.

 

Edited September 24, 2019 at 09:05 AM by Pall
additional information

[Pall]

By the way, in the above example almost all characters are not similar. So called phonetic components, even if they are present in these characters, do not help, obviously. They work only in some limited number of characters used for demonstration of the advantages of the functional analysis. How are you going to remember the pinyin then? Or you will miss it on purpose? The approach proposed above provides learning both pinyin and the meaning, and fast. What else is needed?

[Pall]

Here is a good training. I took a number of randomly found characters and added a set of target characters and mixed up all. In the total number of the characters the share of the target ones is 15%. The task is to find all the target characters, in this case they are again those ten sounding yù,  and recollect their meaning. It's an alternative to flashcards, having certain advantages. When you see a flashcard for the second time, you know that you have seen it before and that you're sopposed to have learnt it to some extent. But in this training you will see many completely uknown characters, and you will not be sure whether you have seen them before or not. I  took words from the HSK6 list to split them into characters, but it can be any set of any complicity. The only limitation, I think, is that this training should be held after the target characters have been learnt a little by other ways.

[Pall]

Sorry, in the above training there are only 8 characters sounding yù, I lost two for some reason while mixing up. But it's even better not to know how many characters of the target group are present in the training set.

[Pall]

When all characters for syllables to which 5 and more characters correspond will be learnt, then those for syllables with 3-4 characters come. But at the next stage , when only those for syllables to which correspond 1-2 characters remain, the learning strategy should be changed. Otherwise it will take a long time to achieve the goal of the 1700 characters.

[Pall]

I have to correct the strategy. It was rather easy to learn characters respresenting syllables to which 5-16 characters correspond, making linear formulas. But there are only 353 such characters, 20% of the total. Characters for syllables, to which correspond 4 characters amount to 216, those with 3 corresponding characters 258. I could create for these lenear formulas, too, but for the characters representing syllables, to which only 3 characters correspond there would be 86 formulas, pretty many compared to 54 for each of the previous cases. And it would be a nightmare to learn the rest - 416 characters, representing syllables, to which only two characters correspond and 530 characters having their own syllable (at least within the HSK5). 

 

Therefore I decided to learn by blocks instead of lines. In the second picture you can see, that 树 is selected as the 'anchor' for the block for syllables starting with 'sh' for the lines where it is and below, and the formula has a dendrit shape. All branches begin with the "tree" and then diverge. In the first picture you can see characters for them. The block includes 17 characters. Numbers in the end of the branches show the number of target words in them, but the branches overlap in their begining. The branch going down includes 9 targert words, but a formular or a branch of a formular should not be longer than 7 target words. Therefore it ends with the 顺, and 顺 is the beginning of its continuation, 3 target words. So, the last word of this branch is the first word of the continuation. The same kind of a link will be for connecting blocks: 树 in its turn will be the end word of the downward branch of the upper block. That will help to observe big parts of the table (for example, for a certain first letter of syllables) in one piece. The same is provided by making an anchor a syllable with the fourth tone (instead of the first one or neutral). The letter option was chosen also due to the fact, that on the average lines for the fourth tone have more characters, it's more popular.

 

 

Edited September 26, 2019 at 05:49 PM by Pall
additional information

[Pall]

Then one should test himself by typing both pinyin and characters in an empty sheet. 

 

When characters are learnt to some extent, the above test can be used, to find all characters with certain pinyin in a mixture with uknown characters (and to recollect the meaning, of course). However, ii's better to include in the mixture some known characters with other pinyin than the searched.

 

 

 

[陳德聰]

I am impressed by the amount of time and effort you have put in to preparing these materials.

[Pall]

I've learnt all the characters for the initial sound 'sh' within the HSK5, 114 of them, and tested my knowledge by finding them and recollecting their pronunciation and meaning in a mixture with other mostly uknown characters, taken from the HSK6 list, twice as big number.

 

The final version of the table is like in the first two pictures of the two sheets, one for characters and the other for the meaning and formulars. 

 

'Sh' is the third largest sound in terms of the number of characters, the first largest is "j" with 160 characters and the second is "y", 140. 

 

There should not be more blocks for an initial sound than 7 for the reason mentioned above, and the less the better in order to provide easier remembering the whole framework for the HSK5. Although in  this particular case with "sh" there are 8 of them, and I couldn't reduce the number. That's because there should be an overall balance. The length of a downward succession in a block should not contain more than one junction (a jump to a quite different formular with the first target word serving as the end word of the above formular), because downward sections are the most difficult to remember due to changing tones and syllables, and the number of branches in a block should not be more than 7 again. In two upper blocks in the "sh" section there are syllables with the fourth tone rejected as anchors (one in each). In the first picture you can see 'shàn' as such. It makes remembering more difficult, of course, since between 'shài' and 'shàng' there are two syllables (without accounting for tones) instead of one in most cases. In addition, the topmost anchor is also for the rest of the lines in the section above the anchor, not only for the below lines. 

 

I projected blocks for the whole end of the pinyin 'alphabet' from "sh" to "zh", seven initial sounds of the 26, corresponding to 635 characters, 36% of the total 1767 (accounting for different pinyin for some characters), and they include 41 blocks. So, there are 15,5 characters per a block in this part, and the total number of blocks for the HSK5 list will be around 120. 

 

For convenience, points of divergence, where the formula splits into branches, are marked with green, and the ends of the branches in violet. Also junctions are marked in blue (not shown in the picture). 

 

When pointing the cursor on the sell left to an anchor on the second sheet, a note appears with the formular for the downword movement to the anchor below, see in the second picture. If pointing on the anchor itself, the formular for the horizon line is shown (the 'shài' has no characters to the right). Pointing at a divergence points will show another continuation of the downward formular, and that at a junction will show up a new formular for the next section, horizontal or vertical. 

 

Also under the brick-color lines corresponding to anchors on the left there are formilars, connecting the next anchor below firstly with this ancor and then with the 'ends' of it branches clockwise, the number shows the number of the branches (5 for 'shài'), see in the first picture. 

 

Numbers in the left upper corner (second picture) show the number of characters in the above section (for 's'): the total, for the characters for syllables to which 5 and more characters correspond, that for 4 characters, 3, 2 and 1. 

 

Important is the formular for the whole section in the third picture, in the bottom left corner. It connects with a sence all the anchors of the section in the downward direction. 

 

Learning the characters takes a form of typing by memory certain blocks on an empty sheet as shown above (not forget type pinyin on the same lines). It can be advised to type from different points: the ancher, the below anchor, one of the ends, a divergence point or a junction. It can be typed from any character after all. For the beginning a standard Windows input facility can be used, which makes the process easier, for it "notices' the succession of typed characters and next time suggest the same characters first. But one can  'deceive' it by changing blocks typed. At the next stage not so advanced input facility should be used like https://www.chinese-tools.com/tools/ime.html  . The position of characters in its lists is always the same. And it would be great, of course, to have a special program, that would mix characters for the same pinyin without tones all the time, before each input. That would be the hardest. Maybe, it appers some day if the proposed learning algorythm becomes popular. 

 

Final test for a section for a certain initial sound can be finding characers in a mixture with other uknown characters. For that I put random two-digit numbers next to the vertical column of characters (fourth picture) and then press 'reorder'. And, of course, one should use the program, a link to which is in one of topics in this forum section, but it would be reasonable only after at least a number of sound sections is learnt. 

 

One should not worry about remembering all the formulars constantly. When he begins typing, formulars will be recollected spontaneously. And with time they will be replaced by a direct association between characters and corresponding sound. Only one thing is necessary to remember initially: the 26 formulars linking anchors for 26 initial sounds. Or one can just remember what fourth tone syllables are selected as anchors for the given initial sound. 

 

I'm uploading the two sheets in the current state. Someone may use my formulars, interpreting them to his native language or English, or create his own. 

 

 

 

 

HSK5 syllables (pinyin&characters) 29.09.19.xlsx

[Pall]

In addition, for better memorizing it would be useful to make for characters  in the sheet 1 notes with showing one of the words created with this character. It's better if the other characters in the word are known already, but it's not a problem if not, they will be known as soon as the section for them is filled in and learnt. 

 

Also one may give a link to a video for this word like the one in the enclosure for the character 沙.

 

Characters used separately can be underlined with two lines.

 

 

[Pall]

Here is another way to fix in one's memory new characters from a part of the table, most difficult. I cannot write in Chinese yet, but I can insert Chinese words in Russian text, and even successions of words. Characters, which are used in words and rearly used as separate characters, I insert in  words. But it's not for learning words, they should be learnt in their use in original texts and speech. That's just to remember the characters better. It's like a diary. Words marked with red include new characters.

29.09.19

晒

我写着 дневник. 闪电 тусклое, сейчас вечер. 可以工作  на  батарее, с более тусклым экраном. Я 善于 в том, чтобы продлить срок службы батареи. Когда ее время работы сокращается, меня это 杀。我会伤心的。 Придется 去商店买    新   电脑。 我不想 быть столь 傻, что работать все время от сети.  Мы не в 沙漠， компьютер всегда можно 买, как 衬衫。 Вы не 欣赏 этой моей способностью? Тогда я все это 删除。 Хотя пусть остается.

 

[Pall]

Why I put so much time and effort into learning the table? It's not only for HSK5, it's also for a more distant future. I have "Dictionary of Contemporary Chinese" for 20,000 words, comprising different fields: everyday life, politics, business, science, etc. And it may be a surprise,  these 20,000 words are based on only 4,500 characters. It's just 2.6 times more than the number of characters in the HSK5.

 

As soon as the table is memorized firmly, it will be easy to add to it new characters (marking them), which will be memorized much faster. I tell my students about the nature of investment. It's exactly that, putting time and effort in 'roundabout processes', portponing consumption and other much needed in order to get  it later in bigger amount and of better quality. The definition is from an American textbook on project analysis. Here are two examples from it.   Investment can be compared to making a net for fishing before fishing. One will be without caught fish for sometime, but afterword there'll be plenty of fish caught. Another example is sharpening the axe before cutting a tree.