Definitions for functional (grammar) Hanzi

[lorentz]

I’m doing the simplified course, so not sure which parts are relevant to the traditional course

Not sure what to call these, but there are many hanzi which have important definitions not really covered by the meaning answer.

1. Particles like “么”
   The current definition is “muh”, presumably because that’s what it sounds like. I think more useful definitions would be “interrogative particle” or “pronoun suffix”, and then you could have note about what these mean. And/or if you want to avoid teaching grammar in the course, I feel like a better answer here would be “me-particle”, and similar for other particles/functional words.

2. Measure words like 本
   The current definition here is notebook. If the goal of the “meaning” response is to help use understand the intrinsic value of the hanzi, I think a better answers would be “root”, “origin”. Another useful alternative would be “book measure word”/“book classifier”, or if you want to enforce a shorthand convention, “book-MW”/“book-CL”. You could further reinforce this by adding 一本书 as a vocab word.

3. Words that have alternate meanings/pronunciation like 还
   The current pronunciation is huan2, and the meaning is “give back”. There’s no mention of the hai2 and the “still”, “even”, “also” meaning. Again, if grammar is out of scope for this app, you could have the suggested “meaning” response be “hai-adverb”.

[kevin]

1. Particles like “么”

My motivation for naming it in this way is indeed to avoid teaching grammar. The meaning “interrogative particle” by itself without any example sentences and in-depth grammatical explanation actually ends up being about as useful as the “muh” we have now, unfortunately, since many people many not be familiar with these grammatical terms. We avoid teaching grammatical terms because they are useful only with ample examples and explanations, which detracts from the core character-learning aspect of the app. I’d love to teach it, but doing so would make learning characters a little bit more work on average, so I have avoided it.

2. Measure words like 本

For each character, we have to balance the following when choosing its definitive meaning:

1. The most common meaning/pronunciation.
2. The relation it has with its “component name” if the character is a single component.
3. The meaning that more logically fits into any words that it forms.

We chose “notebook” here because it helps with (2) and also matches (1) to a degree. I think “book” and the one you mention, “root” are both equally common in practice. For (3) the meaning “root” is a bit more common in words like 根本.

For some characters, like this one, we are going to be torn between focusing on one of the common definitions or the others. I think us having a note mentioning the other may help in this case.

In either case, I noticed that character does not have “book” as an alternate meaning. I’ll add it in the next update. I could see it being a better definitive meaning than “notebook”, for sure. Perhaps we will switch it out in the future.

adding 一本书 as a vocab word.

Adding measure words as vocab words/phrases in this way is a great idea and something we have been thinking about doing.

3. Words that have alternate meanings/pronunciation like 还

A good portion of characters have multiple pronunciations. We try to only teach the most common meaning/pronunciation that is found in vocabulary words. Unfortunately, in some cases the character has two equally common pronunciations/meaning, as in this case. We choose to teach huan2 here because it better fits with vocabulary word meanings.

Right now we do not teach alternate pronunciations for characters, but we have considered adding a guard that would protect against accidentally typing in hai2 in this scenario. Something like “That is valid, but please type in the other meaning/pronunciation”, perhaps. But we haven’t dug too deep into it as of yet.

---

Thanks for the feedback, it is really valuable. For our curriculum, I view it as something we will be continuing to fine-tune, expand, and update every week for years to come. So even if we cannot implement some of these suggestions now due to other things taking priority (for me, it is adding more vocabulary words to keep up with user demand, and fixing up existing mnemonics/meanings), I can see us implementing them in the future.

[lorentz]

We avoid teaching grammatical terms because they are useful only with ample examples and explanations, which detracts from the core character-learning aspect of the app.

Yeah that makes sense. I suppose there are no common terms for these things? I still feel like “-particle” is more useful, cause “muh” has no meaning. It’s not a big deal though, because the user can just put whatever they want as a synonym.

For each character, we have to balance the following when choosing its definitive meaning:

1. The most common meaning/pronunciation.
2. The relation it has with its “component name” if the character is a single component.
3. The meaning that more logically fits into any words that it forms.

The confusing part here is, and this may be me as a long term WK user, is that 1b + 3 would seem to be the most important for long term memory. Also IANAL (I am not a linguist), but I assume that 本 originally meant “root” and associated terms, then abstractly “origin” (in the same way “root” has an abstract meaning in English), and then “book” as some later additional meaning. I’d expect a lot of words to involve the “origin” meaning in some abstract sense. And again this point isn’t a big deal, because we have user added synonyms.

The 一本书 vocab would be super duper dope tho , I think measure words usage in app would be a huge gain over other decks which merely teach HSK word banks.

Along those lines, I hope to eventually see common verbal phrases in the curriculum. As an extreme example of what a mean, in English the words “make” form dozens and dozens of verbal phrases, e.g. “to make sure”, “to make a decision”, etc. In Chinese, I can already see that 打 has this role as a dummy verb, and I’m sure there are more.

A good portion of characters have multiple pronunciations

That’s slightly concerning, How often does this happen? I was always told the beginner trap “unlike Japanese, Chinese characters have only pronunciation except for 了 and maybe a few exceptions”. But now that I’m actually learning, I feel like this happens a lot more than I had anticipated.

[damia]

@kevin

My motivation for naming it in this way is indeed to avoid teaching grammar. The meaning “interrogative particle” by itself without any example sentences and in-depth grammatical explanation actually ends up being about as useful as the “muh” we have now, unfortunately, since many people many not be familiar with these grammatical terms. We avoid teaching grammatical terms because they are useful only with ample examples and explanations, which detracts from the core character-learning aspect of the app. I’d love to teach it, but doing so would make learning characters a little bit more work on average, so I have avoided it.

The thing with this for me is that, while I think it’s good for the course not to teach grammar, students will be learning grammar at the same time anyway (be it with grammar courses or other forms of comprehensible input). And, if you already know the grammar, I feel it is way easier to learn a character relating to that grammar point than adding an additional meaning.

[kevin]

I agree that 1 + 3 is more important in the long run for sure.

As an aside, as I add vocab I go back and tweak these exiting character primary meanings to be more appropriate as they relate to vocab (the #3 mentioned above). Additionally as I do my daily reviews I also tweak the definitions as they relate to #1. You can expect them to get better on both fronts over time. And as you mention, user synonyms can always be used for any spots I miss.

The 一本书 vocab would be super duper dope tho

I’ll keep that in mind!

How often does this happen?

According to this link, about 20% of the time. However, I think this source misses an important point. It will count a character as having “different pronunciation” if it is EVER used a different way. However, that single different pronunciation/usage may be extremely rare - to the point it is not encountered in practice. E.g., it could be a pronunciation only used for rare phrases (chengyu 成語). So I think it is, in practice for a Chinese student/beginner, probably closer to 10% or less.

It is definitely not even close to Japanese, where it seems like characters that are only read a single way are the exception, not the rule.

[kevin]

Can you give me an example of a grammar point and a character that could relate to it? I’m trying to picture what our Lesson could look like to do as you say.

If we are talking about measure words specifically, I can imagine us doing something similar to what @lorentz mentioned, where we teach/mention that a character is a “measure word” and then have that fact reinforced via vocabulary which are just simple measure word phrases. 一本书 one book，三颗黄豆 three soybeans, and so on.

[damia]

I still haven’t done many characters, but an example that stood out recently was 和 being taught as “harmony” instead of “and / together with”.
Other examples could be 呢 being taught as “ni2: woolen cloth” instead of the familiar “…and you?/” or simply “question particle”; 的 being taught as “di4: target” instad of “possessive particle” or “of”; or 會 being taught as “meeting” instead of “can/know how to” meaning or “will”.

[kevin]

Gotcha.

One trade-off here is that if we teach these more grammatical definitions, then we will not be able to teach as many useful vocabulary that align with the current definitions that are more biased towards that end. We view teaching vocabulary important primarily as a way to reinforce the character meanings/pronunciations you previously learn. E.g., knowing 約會 (yue1hui4 “date” or “appointment”) would help the user recall both the pronunciation AND meaning of 會 (hui4 “meeting”).

If we teach more grammatical-focused definitions, then we will need to start teaching extremely short phrases instead. Like “我會說” or “我會去". That is an option, but in this case - and for many grammatical usages of characters - the grammatical definition is extremely large. 會 could mean “know” or “will” or “likely” with equal frequency, for example.

We will keep this in mind as we continue to evolve the curriculum. I wanted to give a background on our thoughts for teaching the particular definitions/pronunciations we do. In some cases like 的 and 呢 I can see us moving towards the more grammatical definitions as they are not ambiguous and more common. Pretty much for each character we have to find a balance when choosing the definition, and with over 3,000 characters it’ll take a while to get perfectly right.

[lorentz]

Throwing my 2cents in here

I think the issue here is how single character words are treated in the curriculum. IMO, there should be a character item and a word item for single characters. The former, as previously discussed should cover the intrinsic meaning the character has, which is to say the meaning it imparts into compound words. The latter should be the meaning of character as a monosyllabic word.

For 和

The character imparts the meaning “harmony”/“peace” into compound words. As for the “and/with” meaning, I’m not sure on the etymology. Wiktionary doesn’t say much, but there is a page on Baidu. Struggling with google translate it looks like the conjunction usage somehow came from the “together” sense of the word. IMO it makes more sense to think of 和 as a word to mean “and”, but as a character to mean “harmony”.

As for 的,

According to wiktionary, one hypothesis is that originally the glyph for this possessive particle was 底. And 的 was later used to write that word instead. So the intrinsic meaning of 的 is indeed “target” (in addition to “bright”, “mark”, and “aim”). But that’s up to the team for which meanings they think are more relevant for the curriculum. One issue for this word is that the super common “possessive particle” meaning is pronounced differently. I’m guessing that one of the tenets of this site is to only teach one pronunciation per character, and it’s not going to work to have a word item 的 pronounced “de”.

WK was forced to make the simplicity vs correctness trade-off because it would be untenable in Japanese to teach only one reading. In the grand scheme of things, it’s not the worst thing for HH to make the other decision - you’re going to learn these exceptions for grammar particles in your grammar studies anyway. I think it’s going to continue to confuse people though.

If I had to TL;DR my feedback here would be

1. Consider using a character/word dichotomy for monosyllabic words (don’t need to mention anything about grammar/particles/measure words in the character definition).

2. Allow “secondary” sound meanings if there is an important vocab which has that meaning.

3. Include measure words as vocab phrases, and for particles add vocab items like “ba-particle” (not mentioning function at all) or the primary meaning (like is the current case for le).

[damia]

I definitely can see the trade-off in teaching the more etymological meanings to understand how the character functions in forming compound words. Still, with such common gramatical characters I still find it distracting to be learning a different meaning and pronunciation as the one you’re seeing all the time, especially in the beginning.

Personally, I don’t see why not teach a primary meaning along with its alternative meanings and later on teach words (or short phrases) that use both. That’s how WaniKani handles kanji with multiple meanings (granted, the problem with basic grammar is way smaller as most of the grammar is made with hiragana). For example, today I unlocked 本人 (“Yourself”) and I feel it would make more sense to learn it from the meaning of “originl” than “book”. Learning these different words from different meanings would help understand the character better to me.

[kevin]

Thank you both @damia and @lorentz for the detailed feedback!

In my view, the main issue really boils down to what @damia says:

Still, with such common gramatical characters I still find it distracting to be learning a different meaning and pronunciation as the one you’re seeing all the time, especially in the beginning.

Namely that the issue is not the grammatical/vocabulary-related distinction, but instead the fact that us being biased towards the latter leads towards us teaching less common pronunciations/meanings for some characters. In other words, a new user only really cares about learning the meaning/pronunciation that is most common above all else - because that is what is ultimately useful.

In this sense teaching 的 as the possessive particle “de5” is much better, because it is the more common meaning and usage and pronunciation. And the use of short grammatical phrases like 我的 ensures that we can still have “vocabulary words” for these alternate pronunciations.

I’m pretty won over by this sort of thinking. I’ll put some time aside this week to think about it more deeply a bit and start making the necessary changes. One tricky thing is that we will need some sort of guard, hint, or notification to users when we change e.g., 的 from di4 to de5. Perhaps just resetting it to be a lesson for everyone once more is the best way.