第041号 · Issue 041
The Pen Is Not the Active Ingredient
{{first_name | みんな}}さん、こんにちは ☕
In 1995, two Japanese researchers ran the experiment everyone in this argument should have read. 📌
Repeated writing improved memory for abstract graphic designs. It did not improve memory for kanji, for words, or for syllables. And in a later experiment in the same paper, repeated writing improved free recall and did nothing at all for recognition.
Thirty years old. Almost nobody cites it. 👀
Know someone with a notebook full of the same character? Send them this. Share this issue →
Two skills, and everyone proves one while claiming the other
Writing a kanji from memory and catching that same kanji inside a sentence are different skills. They have different evidence behind them. 🚪
Nearly every argument about handwriting falls apart in the same place: someone demonstrates the first and claims the second.
So let us keep them apart for the next few minutes.
Writing builds the form, and that part is real. We are not going to soften it. There is a well controlled study where learners who handwrote characters beat both a reading only group and a typing group, on recognising those characters and on linking them to meaning. Two active control groups, matched schedule. It survives most of the objections you could raise at it. Handwriting does do something for recognition. ⚡
Now look at what the other studies compared it against. The famous adult experiment on unfamiliar scripts used passive viewing as its control. Writing versus looking. Of course writing won. 💡
That result tells you retrieval beats staring, and we already knew that. The question you actually have is not "is writing better than nothing." It is "is writing better than twenty minutes of something else," and that study was never built to answer it.
That criticism is ours, by the way. We went looking for someone in the field who had published it and could not find anybody. 🤝
一 言 | ☝️ Chotto ちょっと A 1998 follow up tested whether the benefit came from performing the correct stroke sequence. The results did not support stroke order as the explanation. Uncomfortable, given how most of us were taught. It also points somewhere useful. 🎯 |
So what is doing the work
Most likely retrieval, not the pen.
Which changes what the practice should look like. Covering a character and writing it from memory is a test. Copying a character you can see, twenty times down a page, is not. Those two feel nearly identical while you are doing them, and they are not the same activity. 🎴
危 険 | ⚠️ Abunai 危ない Here is the honest state of it: nobody has run handwriting against equal time retrieval practice for kanji. That experiment does not exist. So the confident people, in both directions, are extrapolating from studies that did not ask their question. Us included, if we push past what is actually here. 😅 |
And the arithmetic only applies to one person: someone writing kanji in order to read better. If you are studying for 漢検, or you write by hand because you like the way it feels, none of this is aimed at you. Enjoying something is not an error that needs correcting. 🙌
This Issue's Japanese
Word
筆記Ordinary, current
Means: writing something down, taking it down in writing. 📌
Use it: 筆記試験a written exam · 筆記用具writing implements · 授業の内容を筆記する。
Reading trap: not ひつき. ヒツ plus キ tightens into ヒッキ. That is gemination, a つ collapsing into a っ. Compare 手書き, where 手 plus 書き voices か into が. That one is rendaku. Two different rules, and a good half of learning readings is noticing which one you are looking at. ⚡
Kanji 漢字
Meaning: brush, pen. The writing tool itself.
Grade: elementary, third year.
Readings: ヒツ · ふで
Common words: 鉛筆pencil · 筆箱pencil case · 筆者the writer · 万年筆fountain pen
The good bit: 鉛筆 is えんぴつ. 万年筆 is まんねんひつ. The same ん sitting in front of the same kanji, two different outcomes. There is no rule you can run to get that one. You know it because you have met both words. 👀
This issue's connection: 筆 is the tool. And the whole argument this issue is having is about how much of the result actually comes from the tool.
Renshuu Time 練習📝
「書くなら、隠して書く」. If you are going to write it, write it covered. About fifteen minutes.
→ Pick three kanji you keep mixing up with something else. Not three you simply do not know yet. The confusable ones.
→ Look at the first one. Cover it. Write it once from memory.
→ Uncover it. Check where it went wrong, not just whether it did.
→ Same for the other two. One rep each, not a page each.
→ Then the half everybody skips. Open something you are already reading, find each of your three kanji in running text, and read those sentences out loud once.
Five minutes writing, ten minutes reading. Most of us have those minutes allocated the other way round. 🎯
You will know you did it when: you hit one of your three kanji in a sentence and you do not stop. Not "I got it right." The other thing, where you are two words further along before you notice which one it was. 💪
✨ Yatta やった
If you only got through the three covered reps and then ran out of evening, that was the part with the evidence behind it. 💡
The tiny win
You are not buying strokes. You are buying attempts from memory. One covered attempt is worth more than a column of twenty copies, and it costs you nineteen fewer.
One reply back
Hit reply with the two kanji you always mix up. We are collecting the pairs, and the ones that keep coming back are going into a future issue. 🍵
Stay connected 🙌
We are around during the week if you want to talk shop, share a win, or just lurk. Both are completely fine.
Follow us: X / Twitter · Reddit
これからも一緒に頑張りましょうね〜 🍵
Kotoba Brew Editorial Desk