ChristopherFritz's Study Log

[ChristopherFritz]

November Day 28

Back around the end of 2014, Cure Dolly wrote about what she calls “Sound Sisters”.

Sound Sisters are similar to WaniKani’s mnemonic system.

The key points for Sound Sisters (and where they differ from WaniKani) are:

1. They focus only on the reading of kanji (with no regard to the meaning).
2. Different kanji with the same reading use different mnemonics.
3. Different mnemonics for the same reading are grouped together.

WaniKani uses the same mnemonic for all matching readings.

In WaniKani, 司 uses sheep as its reading mnemonic. The kanji 士 also uses sheep. And while the kanji 次 is given Jesus as a reading mnemonic, 姿, 諮, and 資 all use sheep.

There’s nothing wrong with this. But for me personally, it’s too vague. I often can’t remember the mnemonic to get to the sound.

I’ve always struggled with WaniKani’s mnemonics.

I can create my own, but whatever mnemonic I come up with for a sound for one kanji may not work out for me for kanji I won’t learn until later. That results in having to concoct absurd stories, which WaniKani states to be a plus (and I don’t doubt that), but the concept of making stories absurd to be easier to remember doesn’t work for me personally.

Sound Sisters uses different mnemonics based on the kanji.

The kanji 司, 士, and 次 are each given their own mnemonic. As a group, the mnemonics are called “She-Knights”, because they share the reading シ.

@Cure Dolly

The first Shi sister is a she-samurai 士.

The second Shi sister 司 wears a knightly helmet with a mouthpiece and eye-slit (the left side is left open so she can get into it).

The last Shi sister 次 is not one actually of the knights but their dependable follower.

It’s very easy to think of a trio of characters involving a sword-carrying warrior, a helmet-wearing knight, and a companion. I mean, such groups have been popular since Ancient Greece:

(Even Gabrielle, Warrior Princess Xena, and Joxer the Mighty are in awe as they look up at the quoted mnemonic.)

Tying a sound to a kanji allows you to utilize it in more complex kanji.

Once you have in place that the kanji 司 is read as シ, it helps with reading words such as 伺候, 歌詞, and 飼料.

I recently learned 候 in WaniKani. Although WaniKani doesn’t teach 伺 with the reading シ, I’m able now to read 伺候 as しこう. I won’t learn 飼 in WaniKani for a while, but I do know 料, so now I can read 飼料 as しりょう.

Granted, there are kanji that include 司, 士, or 次 and are not pronounced シ. 売店 is one example. I’ll tackle those as I encounter them.

As a fiction writer, creating unique individual people that are very memorable to me is a strength of mine.

I plan to take Cure Dolly’s “Sound Sisters” Anki deck, mine the kanji out of it (already done), group them based on their Sound Sister kanji (already done), and look up WaniKani vocabulary that use those kanji plus their meanings (in progress).

By pulling together the meanings of the WaniKani words, I can create specific characters without fear that they won’t work for kanji I learn later. (This won’t help me for kanji not taught by WaniKani, but I’ll cross that bridge when I come to it in a few years.)

Samurai 士: The kanji 士 appears in words involving work, in words involving intent and will, and at the end of various job titles. It’s also used in magazine. Thus, I picture the samurai as a Jack Jill-of-all-trades type. A hard worker, who will take on any role to help out others, whether it means being a soldier, an attorney, or a shogi player. Her flaw is she tends to switch roles based on what she sees in the magazines she’s always reading.

Helmet Knight 司: The kanji 司 appears in words including boss, sushi, grammar terms, and domestication of animals. Combing the her top grammar, her nurturing experience with animals, and how often she buys sushi for the group, it’s no wonder the knight became the group’s boss.

Dependable Follower 次: Aside from 次 meaning next, it’s also used in words related to figure and posture, as well as table of contents, consult, and resources. She may not be a warrior, but she is a vital member of the group as she manages all their resources. She holes the strings of the purse, and the others must consult with her on what they can and cannot buy. She is considered an esteemed member of the group, so she is able to stand stall and proud.

I worked out the details for my version of the “She-Knights” on Thursday, so I won’t go over any this weekend (gotta get in some reading!), but in December I plan hope to plan to go over a different “Sound Sister” group each weekend.

The Sound Sisters only cover 289 of WaniKani’s 2,055 kanji.

That sounds like a small amount. Clearly I’m not getting a huge boost WaniKani from this.

However, in the past couple of days, I’ve already encountered a few reviews that I would have gotten wrong had I not noticed a “She-Knights” group kanji in the word.

As it turns out, the Sound Sisters cover 14% of the kanji covered by WaniKani. I don’t know how many of the individual WaniKani vocabulary words this covers, so if we imagine it’s also about 14% (just for sake of argument), then that means if I have 70 vocabulary reviews in a day, about 10 of those may be Sound Sisters.

In all, I’ll have to create 92 characters for the whole Sound Sisters Anki deck’s material.

That’s somewhere between Fire Emblem cast levels and Musuo cast level.

For now, I’ll focus on the first 6 groups Cure Dolly introduces, which covers 22 main kanji, and another 98 kanji that incorporate them.

[Rowena]

From what you’ve written above, you might be interested in reading The Kanji Code as it deals with both the phonetic components and the semantics, meaning there is less need for mnemonics.

I also look up all new radicals and kanji in an etymological dictionary right after their lessons, which makes more sense of them than some crazy mnemonic.

I write all this because since I started with the dictionary habit, my leech count has been between 40 to 70, so I’m hoping this step-around of the WK mnemonics might work better for you too.

[ChristopherFritz]

The sample page photos on the site aren’t enough for me to make a firm purchase decision just yet, but I’ll keep it in mind. Thanks for the recommendation!

In the meantime, as I do more kanji lessons, and as I get other kanji as high level leeches, I should probably look them up in my copy of “The Complete Guide to Japanese Kanji”. I failed on reading it from the start because it was boring and nothing stuck that way (some years back), so maybe it’ll work using it as a WaniKani lesson/leech supplement.

[Arzar33]

@ChristopherFritz

I plan to take Cure Dolly’s “Sound Sisters” Anki deck

@ChristopherFritz

I should probably look them up in my copy of “The Complete Guide to Japanese Kanji”.

Why use inferior solution when the glorious Keisei 形声 Semantic-Phonetic Composition script already exist

I unironically think that the Semantic-Phonetic Composition script alone make WK the best way to study kanji out there !

[Elizabella]

@Rowena

The Kanji Code as it deals with both the phonetic components and the semantics, meaning there is less need for mnemonics

I have seen this book mentioned in the forums before! I’m assuming that you find it a helpful resource in terms of identifying patterns within the kanji, but you’re using it in conjunction with an etymological dictionary, or you switch back and forth?

@ChristopherFritz

my copy of “The Complete Guide to Japanese Kanji”. I failed on reading it from the start because it was boring and nothing stuck that way

I checked this out of the library awhile back and I just couldn’t get into it either. I mean as a research text I can see its value, but it didn’t feel like a must buy for everyday use.

[Rowena]

@Elizabella

you’re using it in conjunction with an etymological dictionary, or you switch back and forth

I use the dictionary after lessons each day to look up those specific kanji/radicals - it often goes further back into the origins of how the kanji came to be (sometimes that includes mis-transcribing of old characters leading to their modern versions). I am reading through the book to get a better overall sense of the patterns to the readings and semantics.

@Arzar33

Why use inferior solution when the glorious Keisei 形声 Semantic-Phonetic Composition script already exist

I do have this running, and it has helped, but I frankly don’t quite get what these symbols are about

and have not found an explanation anywhere…(I mean, I get that the sound is different, but not an explanation of why, if it is predictable, etc., and the book seems to offer more on that).

[ChristopherFritz]

@Arzar33

Why use inferior solution when the glorious Keisei 形声 Semantic-Phonetic Composition script already exist

I have it installed, but 99.9% of the time I do reviews via the Flaming Durtles smartphone app. The times that I do do reviews on my desktop computer, when I look at the results of the script, often I can’t make heads nor tails of it. Maybe I’m just seeing it on all the worst kanji for it, though.

Perhaps that should be the first place I look after a kanji lesson, then in the Guide book I have (because I already own it), and see how that goes. But also reference the data I extracted from the Sound Sisters Anki deck to see how that lays things out. Doing that for lessons (and for at least one leech kanji per day) may give me an idea of whether any of these resources can bolster my kanji learning.

@Rowena

but I frankly don’t quite get what these symbols are about

I have the same issue. But looking at a few kanji just now, I see:

• From 訟：「The phonetic mark is no indicator at all（下）for the readings of this kanji, they may have changed over time or the composition of this character is different after all.」
• From 影：「The phonetic mark is a good match（上）because the main readings of this kanji are derived from it.」
• From 効：「The phonetic mark is a perfect match（天）because all readings of this kanji are derived from it.」

I’m not seeing any with a note for 中 after checking about ten kanji. Maybe if I check the script’s source code… Looks like there’s no sentence for 中 like the ones above.

Oh, here we go!

@acm2010

I should really add the explanation for the quality levels somewhere

天 means that all readings of a compound match
上 means that the main reading matches, but you can also read it in a way not covered by the tone mark, but rarely
中 means that the main reading is different (かん here), but at least one of the phonetic readings is also used as a reading for the kanji (けん), just not in first place
下 means that the phonetic readings are nowhere to be found.

[Rowena]

@ChristopherFritz

Oh, here we go!

@acm2010

I should really add the explanation for the quality levels somewhere

天 means that all readings of a compound match
上 means that the main reading matches, but you can also read it in a way not covered by the tone mark, but rarely
中 means that the main reading is different (かん here), but at least one of the phonetic readings is also used as a reading for the kanji (けん), just not in first place
下 means that the phonetic readings are nowhere to be found.

Ahh, I’m blind!! I had eventually figured that much out, though not the patterns behind why, thus the book. Thanks so much for digging that up and posting it here.

[ChristopherFritz]

@Arzar33

I unironically think that the Semantic-Phonetic Composition script alone make WK the best way to study kanji out there !

My kanji lesson for today shows why this script confused me enough into mostly ignoring it before:

I’m not quite certain where 礼 and 体 come into play with 豊.

[Arzar33]

Confusing indeed

In those cases, it’s almost always due to historical reasons and can be safely skipped. Here it’s because the 旧字体 of 礼 is 禮 and the 旧字体 of 体 is 體. Even knowing that, the onyomi are so different it’s quite useless.

In general, phonetic components are by no mean a silver bullet. There is a lot of case where the phonetic part is unclear or not useful and can be skipped. In fact according to this article, there is only between 20% to 50% of kanji that are phonetic, depending how lose we define “phonetic”.

It still a big help IMHO. Also the phonetic component increase in efficiency with the levels. At level 25, I think it barely starts to become useful, because we just don’t know enough kanji yet to notice all the connections.

I was curious to know how much phonetic hint can be really used, so I did an analysis of all the kanji of level 25. Here it is:

抜, 鮮, 満, 与, 掛, 隠, 模, 含, 替, 訴, 訟, 限, 豊, 況, 逮, 属, 響 : No clear phonetic part or too confusing for me.

• 渡 と close to 度 ど
• 慣 かん same as 貫
• 候 こう same as 喉
• 捜 そう same as 痩
• 居 きょ same as 据 . Something important to realize here is that 居 itself is composed with the phonetic component 古(こ) (base for 枯, 故, 固, 胡). So 居 almost match its phonetic component, but not completely, and become itself a base. It happens quite often.
• 輩 is part of the big 非 serie, which is a frustrating one. Can be either ひ or はい 緋, 輩, 扉, 悲, 排, 俳, 斐
• 模 part of the weird 莫 serie that can be either ばく/まく or ぼ：膜, 模, 漠, 幕, 墓, 募
• 票 ひょう base for 標 and 漂
• 捉 part of the 足 (そく) serie 捉, 促
• 効 is part of the 交 (こう) serie 絞, 郊, 校
• 景 けい base for the serie 憬 and 影
• 巻 part of the 龹 serie which is either けん or かん 券, 拳, 圏
• 構 part of the perfect serie 冓 こう: 購, 講, 溝
• 補 is a jackpot. It really pay off to remember 甫 as ほ, because it appear in a lot of kanji 哺, 輔, 舗, 補, 浦, 捕

Results:
17 kanji without clear phonetic
14 kanji with phonetic hint

[Jonapedia]

@Arzar33

there is only between 20% to 50% of kanji that are phonetic, depending how lose we define “phonetic”.

I think that, as you have pointed out, knowing phonetic components are useful, but they’re not universal. This isn’t surprising though, given that there are six primary mechanisms of kanji formation (六書):
象形 – pictograms i.e. kanji that look like what they represent
指事 – symbolic i.e. kanji that use symbols/markings to refer to an abstract concept (my dictionary’s translation is ‘self-explanatory’, but I think that helps no one without background knowledge)
会意 – associated meaning i.e. when kanji are combined in order to bring their meanings together
形声 – pictophonetic i.e. kanji that contain a phonetic element and a meaning element (which is what you discussed)
転注 – mutually synonymous i.e. kanji whose meaning has evolved or whose meaning overlaps with that of other kanji that look similar
仮借 – phonetic loans i.e. kanji who are used purely for their pronunciation because a kanji didn’t originally exist to represent the idea (kinda like 当て字 in Japanese)

@Arzar33

抜, 鮮, 満, 与, 掛, 隠, 模, 含, 替, 訴, 訟, 限, 豊, 況, 逮, 属, 響 : No clear phonetic part or too confusing for me.

The ones whose phonetic components I can spot:
満 – it’s the thing on the right. It’s related to kanji like 瞞 (also まん), which I know in Chinese, but have never seen in Japanese
掛 – 卦 (か・け)
隠 – purely hypothetical, but I think the little bit in the top right-hand corner has something to do with it. You ever see characters like 淫 (いん)? Also, it might be worth examining the traditional form: 隱. I think the similarity is more obvious.
模 – 莫. Even in Chinese though, this gives us plenty of readings, in particular mo and mu (with various tones too), so it’s not that helpful. It just gives you an idea of roughly what sounds might fit.
況 – 兄. It works a lot better with on’yomi in Japanese, because both are きょう. In Mandarin, the former is kuàng, and the latter is xiōng.
逮 – almost definitely not phonetic. The component above 辶 traditionally represented people like servants or slaves. You find it on the right-hand side of the second kanji in 奴隷 (slave).
響 – 郷 (ごう). This is a case where Mandarin is more consistent: the former is xiǎng; the latter is xiāng. However, it’s not uncommon to find きょう・こう・ごう being associated in the on’yomi of related kanji.

[ChristopherFritz]

@Arzar33

At level 25, I think it barely starts to become useful, because we just don’t know enough kanji yet to notice all the connections.

That sounds about right for why I haven’t noticed it too heavily yet. I mean, there are certainly times where I’ve seen one kanji uses another kanji within it and they have the same reading. I kept feeling like there must be something I’m missing out on.

Cure Dolly’s Sound Sisters deck covers only 163 kanji for the first 25 WK levels (and only 218 kanji for the last 35 WK levels), so it won’t get me quite as far as I was hoping. (I realize it’s not a cure-all solution.) And as luck would have it, none of my kanji leeches are in the deck. I’ll still keep an eye on the Sound Sisters deck material, but right now I’m left a bit open on what I’ll focus on in December (aside from reading).

@Arzar33

14 kanji with phonetic hint

This is what I need to pay more attention to. I wonder if anyone’s put together a chart that groups all WK kanji similar to what you listed.

[ChristopherFritz]

As 2020 nears an end, how are my goals for the year looking?

I follow the Scott Adams perspective: systems are better than goals. For that reason alone, missing a goal is okay so long as I’ve kept up a system toward the goal.

@ChristopherFritz

Reading Goals

1. Read 30 manga volumes.

A system of reading each day has gone very well for me, up until I stopped following the system.

My reading’s really tapered off lately. A big part of that has been playing the latest Hyrule Warriors game. As I near completing it I really need to get back into a habit of reading. I thought removing the number one friction to reading (tracking daily page counts) would lead to more reading, but I’ve been reading a lot less this past month.

That said, here’s how I did on the series I expected to read at least 23 volumes of:

• 怪盗セイント・テール: Read (past tense) the final volume.
• よつばと！: Read the final five (available) volume.
• 三ツ星カラーズ: Read the last five (available) volumes.
• アクア/アリア: Completed three volumes. I considered setting it aside and loading vocabulary from the anime into Anki, only to find there are no Japanese subtitles available =(
• ふらいんぐうぃっち: Six volumes completed. Will complete a seventh before December is over.
• 美少女戦士セーラームーン: Not counting side stories, I’ve completed eight kanzenban volumes.
• GALS！: Completed one volume.

That comes out to 29, meaning I almost reached the 30 volume goal there alone.

However, I already read other series I didn’t originally plan on:

• アオハライド: Will have completed eight volumes by the end of December.
• おじさまと猫: Completed one volume.
• GOSICK―ゴシック―: (Sort of) completed one volume.
• ハヤテのごとく！: (Barely) completed one volume.
• レンタルおにいちゃん: Read all four volumes.
• 魔女の宅急便 (cinemanga): Read whole movie.
• 異国迷路のクロワーゼ: Completed one volume.

By the end of this month, that’ll be another 17 volumes, putting me at a total of 46 volumes read. Looks like I may have to aim for 50 volumes in 2021! Although that will be harder without the easy ones…

@ChristopherFritz

2. Read four children’s detective picture stories. Not very ambitious, but my focus is on manga.

I planned on reading the three わんわん探偵団 books, but since we have the わんにゃん book club, I instead only read the first one, with the club. (Next book is slated for January.)

I also planned on reading the ミルキー detective series from the same author. I’ve completed four books from that series.

Total read: 5 books, putting me above goal. I feel like I should increase this number for 2021. There are 9 more books in the ミルキー series if I don’t count that later books that mysteriously never received e-book releases. I may just have to buy them paperback when I get to them…

@ChristopherFritz

3. Read a Studio Ghibli シネマコミック (cinemanga). I read through となりのトトロ in 2019.

As mentioned above, I read through the 魔女の宅急便 cinemanga. I should go through at least one more in 2021. Not sure which, as I’d like to tackle the original Whisper of the Heart manga sometime before going with the Ghibli release.

@ChristopherFritz

Vocabulary Goals

Create a custom deck in iKnow where I add at least one Core vocabulary word I encounter in reading each day.

This will require I get my backlog of reviews down to zero.

iKnow’s interface made this too cumbersome. I’m toying with starting an Anki deck, adding some words I look up while reading manga. Just need to make sure I’m not adding words I’m unlikely to see again any time soon. (I’ll probably do an anime subtitle search to see which candidates get a lot off hits.) But I also don’t want to waste time reviewing words with kanji I won’t recognize until I learn the kanji in WaniKani a year later.

And I’ve been doing a really good job of getting my WaniKani reviews down to zero (almost) every night.

@ChristopherFritz

Grammar Goals

Identify at least one grammar point I’m certain I don’t know (or don’t know well) in what I’m reading, and thoroughly read up on it.

I’ve found this doesn’t system really work for me anymore. I’m not good at identifying what I need to learn better. But participating in book clubs does help with this!

@ChristopherFritz

WaniKani Goals

Reach level 40. Hopefully this is setting the bar low, but going up another 23 levels in the same time I did 15 sounds like a lot to me. I know I slacked off on lessons and didn’t get serious about taking on more lessons for much of the first half of 2019.

If I was 23 levels away from 40, then I was on about level 17. I’ll hit level 26 in a few days, meaning I didn’t even make it 10 levels in 2020. In my defense, I didn’t know about the wonderful world of #NoLeechNoLife a year ago!

A big part of the slow-down was to keep my apprentice count around 100 (and more recently around 50).

Aside from that, I’ve also leaned more heavily into reading than I expected to, plus I’ve increased my book club discussion participation a bit.

Missing the goal (whether by a little or a whole lot) is okay, because the system is to continuously learn new kanji, and I’ve succeeded in doing that.

My goal for the end of 2021 is to reach 35, meaning 10 levels.

@ChristopherFritz

Speaking Goals

I have no reason to learn to speak Japanese, but I wonder if shadowing will help my listening speed and comprehension speed any. I’ve been planning to shadow 「となりのトトロ」 in 2020 (but not right away). Since I know this movie very well, and have seen it probably 30 to 40 times (although most of that in English), it should be a good source to try with.

Nah, I didn’t get around to that.

[tommyboi]

@ChristopherFritz

A big part of that has been playing the latest Hyrule Warriors game.

How is it? I’ve never played a Warriors game but loved BOTW. Decisions, decisions…

@ChristopherFritz

I’m toying with starting an Anki deck

I’ve never used iKnow, but Anki is pretty chill once you wrap your head around it. It’s also super easy to delete cards while you’re reviewing if you later decide it’s not worth adding. You can also ‘suspend’ cards so that you won’t see it in your reviews until you unsuspend it later. I agree with the Kanji point though, my retention is pretty laugable for kanji I haven’t seen in WaniKani. At this point I’m using it more for exposure of new words and grammar structures.

@ChristopherFritz

I’ve increased my book club discussion participation a bit.

a bit - Understatement of the century lmao

Congratulations on your progress this year!

[ChristopherFritz]

@tommyboi

How is it? I’ve never played a Warriors game but loved BOTW. Decisions, decisions…

If you’ve ever played a Warrio–oh, well. Never mind that.

I think if you’ve never played a Warriors game, then it’s worth trying one out. I wasn’t interested in a single Warriors game before (the first) Hyrule Warriors. And I’m still not interested in any others. The grind would be too much for characters I don’t care about.

I think the gamplay in Age of Calamity has improved upon the first Hyrule Warriors in just about every way except that the first Hyrule Warriors had more complex missions. I feel like they may have dumbed them down a wee bit to aim for a wider audience knowing there’d be people picking up this game for more Breath of the Wild storyline. I still like it, though.

If you’re interested in the backstory to Breath of the Wild, and would like to visit that, then download the Age of Calamity demo and try it out. See if you like the gameplay well enough. Main storyline will likely take about 30 hours to play through for a beginner to the Warriors genre.

The main complaint for Age of Calamity as telling the backstory is that (as per the very first scene in the game, so I’ll not mark it as a spoiler) it’s an alternate storyline. An event at the very beginning branches the events off in a different direction from Breath of the Wild’s history.

It’s kind of like how Ocarina of Time (if you’ve played it) is a “bad ending” (Ganondorf takes over), and then when you beat the game, Zelda sends Link back in time to stop Ganondorf, preventing more than 95% of the game’s events from ever taking place. Likewise, Age of Calamity is a time travel adjustment to try and prevent what leads to Breath of the Wild’s.

@tommyboi

I’ve never used iKnow, but Anki is pretty chill once you wrap your head around it.

I’ve used Anki a few times before, most notably for learning all of (and since forgetting some of) the words in volume 1 of ごちうさ in 2018.

Main problem was never knowing how to use it well. Any way I tried to use it often resulted in me hitting a limitation of the software.

If I do use Anki, I’ll use it with cards made along the lines of what Cure Dolly suggests, and using options along the lines of what refold.la recommends.

@tommyboi

a bit - Understatement of the century lmao

When I first joined the ABBC last year, I was really uncertain of what kind of discussion would (or should) take place. I tried doing various different things, and along the way encountered some pushback on doing posts that covered too much material. From there I adjusted to doing more detailed replies of questions asked by others, and that seems to be working.

For the レンタルおにいちゃん book club, I tend to bounce back and forth between wanting to get people responses as soon as possible, and wanting to let others respond. One might ask, what’s the difference, is the response will be more or less the same? Thing is, answering questions is one of the best ways to either learn material better, or solidify one’s own understanding of the grammar involved. In other words, every time I respond, I may be robbing someone else of the chance to improve a little bit!

[tommyboi]

@ChristopherFritz

it’s an alternate storyline

Oooh interesting, I was wondering how the game would work given the backdrop in BOTW. I also didn’t realize they made a demo, I’ll have to try it out

@ChristopherFritz

I’ll use it with cards made along the lines of what Cure Dolly suggests

I’ll look into this, they recommend standalone vocab right? I’ve been doing sentence cards so far which I really liked at the beginning, but I can feel my motivation waning. I think if I mined my own sentences it might be more interesting, but I don’t seem to have the motivation to do that. Also, I seem to have pretty good retention with WaniKani vocab so maybe it’s time for me to reevaluate.

@ChristopherFritz

I may be robbing someone else of the chance to improve a little bit!

I see what you’re saying, but I for one am very greatful of your explanations!

[ChristopherFritz]

@tommyboi

I’ll look into this, they recommend standalone vocab right?

Correct. And anything that helps you learn/remember the word can go onto the back.

I did sentence cards with my ごちうさ deck in 2018, where a sentence appeared on the front with the word in review being highlighted. Problem was, often I would remember the word based on the other words in the sentence.

I’d seen someone ask about exactly this as an issue on reddit’s LearnJapanese board, and the common response was that this is a good thing. You want to be able to recognize the word in context. So I accepted it. Fast-forward to when I started doing more reading in 2019, and I found I didn’t recognize those words one bit when I saw them in other manga. Turns out that recognizing a word because of the sentence on the front of the card meant I didn’t need to even pay attention to the word’s kanji. Big fail.

Reviewing words in isolation feels wrong to me, but it’s sort of worked(ish) with for me with WaniKani. I will absolutely have sentences on the backs of the cards, though, as I’d be “mining” words from the manga I’m reading. So, I’m reviewing the card with no context, but I’ll still be seeing the context sentence on the back each time.

@tommyboi

I’ve been doing sentence cards so far which I really liked at the beginning, but I can feel my motivation waning.

If you don’t know the sentence from someone (such as a manga you’re reading), it’s a bit hard to really get into it, isn’t it? One thing I did like about my ごちうさ deck was that seeing the sentence let me visualize the scene it’s from.

I’m considering putting whole manga panels into this proposed deck. The main thing there is that Anki (as far as I know) doesn’t let a person synchronize decks between two devices directly. They have to synchronize through Anki’s web site, meaning there are limitations on the overall file size (whatever that may be).

Cure Dolly also recommends having audio for the words, even if it’s via TTS (text to speech). It’s something I’ll want to look into as well, as there are ways to automate the process of adding audio.

@tommyboi

I think if I mined my own sentences it might be more interesting, but I don’t seem to have the motivation to do that.

If you can type up a sentence to ask about it in a book club, you can type a word and a sentence into a note in Anki! =D

I struggle most with what to put onto a card, so I’m going to re-watch Cure Dolly’s videos on using Anki sometime, and if I don’t get a solid answer there, I’ll go with what refold.la has:

I may split reading and meaning into two separate fields, and add a generic notes field.

[Elizabella]

@ChristopherFritz

systems are better than goals. For that reason alone, missing a goal is okay so long as I’ve kept up a system toward the goal

Well this makes all kinds of sense… now to think about what systems I want to set up.

[ChristopherFritz]

@Elizabella

Well this makes all kinds of sense… now to think about what systems I want to set up.

Here’s how Scott Adams puts it:

@Scott Adams

Going to the gym 3-4 times a week is a goal. And it can be a hard one to accomplish for people who don’t enjoy exercise. Exercising 3-4 times a week can feel like punishment – especially if you overdo it because you’re impatient to get results. When you associate discomfort with exercise you inadvertently train yourself to stop doing it. Eventually you will find yourself “too busy” to keep up your 3-4 days of exercise. The real reason will be because it just hurts and you don’t want to do it anymore. And if you do manage to stay with your goal, you use up your limited supply of willpower.

He talks more about willpower in the whole post.

@Scott Adams

Compare the goal of exercising 3-4 times a week with a system of being active every day at a level that feels good, while continuously learning about the best methods of exercise. Before long your body will be trained, like Pavlov’s dogs, to crave the psychological lift you get from being active every day. It will soon become easier to exercise than to skip it – no willpower required. And your natural inclination for challenge and variety will gently nudge you toward higher levels of daily activity while at the same time you are learning in your spare time how to exercise in the most effective way. That’s a system.

I’m working on my 2021 reading goals. Sounds counter-intuitive, to be working on goals while talking about systems being better, right? That’s because the goals are just for fun.

My systems toward reading in 2020 were:

1. Learn more grammar.
   • Watching Cure Dolly videos daily for much of the year has helped a lot.
   • Answering questions in book clubs discussions has required me to learn grammar better.
2. Join more book clubs.
   • I started a Sailormoon book club.
   • I joined the Aoharu book club.
   • I’ve continued the Flying Witch book club.
   • I continued the Yotsuba book club.

If I simply focused on reading and nothing else (except WaniKani), I would not have completed 30 volumes this year.

Keeping up a routine helps me as well, especially if it’s a daily task. My routine broke back around November (for various reasons), but my book club routine’s mostly held up.

Setting reading goals for 2021 will be about working out my daily reading routine. (Scott is positive on having a daily routine as part of one’s system.)

Back to Scott, he used to write blog posts. Writing the posts doubled his daily workload (for however long his main job doing a daily newspaper comic takes in a day), yet didn’t bring in nearly as much income as the comic strip. He maintained that writing blog posts was a system:

@Scott Adams

Writing is a skill that requires practice. So the first part of my system involves practicing on a regular basis. I didn’t know what I was practicing for, exactly, and that’s what makes it a system and not a goal. I was moving from a place with low odds (being an out-of-practice writer) to a place of good odds (a well-practiced writer with higher visibility).

This is something applicable to life in general. If Scott is an out-of-practice writer, he won’t be getting any writing jobs. He didn’t have a specific writing jobs in mind that he was aiming for. He simply was increasing his chances of landing any kind of writing job.

Scott ended up getting paid to write articles for the Wall Street Journal. Not much in the way of money, but that lead to book publishers noticing him. And the book he wrote lead to speaking deals, which can be quite lucrative.

He didn’t write blog posts to one day get speaking engagements. And if his goal was to only write blog posts, and if he turned down the Wall Street Journal, he wouldn’t have written a book, and he wouldn’t have found himself speaking before large crowds.

@Scott Adams

My problem with goals is that they are limiting. Granted, if you focus on one particular goal, your odds of achieving it are better than if you have no goal. But you also miss out on opportunities that might have been far better than your goal. Systems, however, simply move you from a game with low odds to a game with better odds. With a system you are less likely to miss one opportunity because you were too focused on another. With a system, you are always scanning for any opportunity.

My systems in 2020 didn’t really move me beyond my goal, except that I read more than planned.

However, I’ve found recently that my ability to understand anime in Japanese (for shows I’ve seen before, and have playing in the background) has increased this year. I’m able to catch more words and follow along better that I did a year ago. If I focused on reading and didn’t improve my grammar as much, I think this extra gain may not have taken place (or at least not as much as it has).

I haven’t worked out what my systems for 2021 will be beyond more of the same from 2020. It’s getting harder though as I’ve learned a lot of the “basic” grammar that comes up most commonly in manga.

[ChristopherFritz]

@tommyboi

I’ve been doing sentence cards so far which I really liked at the beginning, but I can feel my motivation waning. I think if I mined my own sentences it might be more interesting, but I don’t seem to have the motivation to do that.

@ChristopherFritz

Reviewing words in isolation feels wrong to me, but it’s sort of worked(ish) with for me with WaniKani. I will absolutely have sentences on the backs of the cards, though, as I’d be “mining” words from the manga I’m reading. So, I’m reviewing the card with no context, but I’ll still be seeing the context sentence on the back each time.

I’ve been reading refold.la a little here and there this past week. It’s very well written, and a lot of what they say about the experience of learning a language matches my own. I don’t think they are necessarily infallible, but it’s written by one or more people with more experience and, more notably, more success than me in language learning. Thus, I give some weight to what they write.

I reached their page on sentence mining last night.

They recommend putting a sentence on the front of the card.

Just as putting one word alone on the front feels wrong to me, I know I’ve had a bad experience with sentences where I learn the word in context of the sentence, but not the word itself. This was before I started actually learning kanji, however. It’s possible I can improve the experience of having a sentence on the front.

The whole page I linked to is worth a read for anyone who has done or is considering doing sentence cards. Here are their main points:

@refold.la

1. Prioritize words that feel familiar.

Words that occur frequently in your immersion provide a high return on investment. If a word seems to be drawing your attention or feels familiar to you, that’s a good indication that it’s a common word.

2. Choose words that seem accessible.

If you look up a word and the definition feels hard to grasp, then skip it for now. Your time and efforts are better spent learning words that are already within reach.

3. Prioritize words that you want to learn.

If a word is interesting and just seems to be calling out to you, that’s reason enough to learn it. An emotional connection to a word makes it much easier to memorize, reducing the “cost” of learning it. This can make learning a word worth it, even if it doesn’t seem particularly useful or frequently used.

These fall under the category of “low hanging fruit”. I think this may be the way for me to go forward.

I’m just not able to do 20 lessons per day of random new words, even one day let alone every day, like some people can. My memory wasn’t that good when they’d drop 50 vocabulary words on us Monday morning in high school English class and expect us to know them all by Friday, and it’s certainly no better decades later.

But there are certainly “low hanging fruit” words that I encounter all the time when reading manga. Words that feel familiar because I’ve seen them before. Words that are easy to grasp. Words that I want to learn as I see them while reading.

I still need to re-review Cure Dolly’s Anki videos to see her recommendation. Whether I put a sentence on the front or back is still undecided. But I feel I have a better idea of what to make a card out of now.

TL;DR: Add cards to Anki through sentence mining when you encounter words that either feel familiar, or are easy to understand, or you want to learn.