What's the deal with/origin of the character 曰?

Question

I'm talking about the 曰 from 曰【いわ】く, not the common 日【ひ】 we all know and love.

1. Why would they "make" two characters that look (for all intents and purposes) exactly the same?
2. How do you really differentiate them except by context? 曰 just looks like a ずんぐりした 日. Presumably the stroke order is the same? Because I haven't been able to find it anywhere. What would you have to do when actually writing it to make sure it's not mistakable? It looks like the middle line of 曰 intentionally doesn't go all the way across.

I simultaneously love and hate this character. Any insight into its mysterious existence is appreciated.

Answer 1

Why would they "make" two characters that look (for all intents and purposes) exactly the same?

They didn't look the same, the shapes just happened to converge.「日」depicts the sun:

商
甲

前4.29.5
合集8236
篆

說文解字
　現代
楷

標楷體
　現代
楷

EPSON正楷
　

The horizontal line in the middle (sometimes a dot, sometimes completely absent) was to distinguish「日」from visually similar shapes such as「囗」and「〇」.

「口」(mouth) was originally not confusable with these shapes, being a picture of a mouth with the corners of the mouth obviously drawn:

商
甲

珠579
合集27706秦
簡

睡ㆍ為32
　現代
楷

標楷體
　

Unfortunately, the corners of the mouth later disappeared, leaving us with a simple shape commonly confused with「囗」(to surround) and「〇」(a circle) in the modern script.

The character「曰」was created from「口」, being originally a depiction of a mouth with a mark above it representing outwards direction from the mouth, indicating the meaning to speak.

商
甲

前7.17.4
合集8233秦
簡

睡ㆍ法121
　現代
楷

標楷體
　現代
楷

EPSON正楷
　

Adding a small mark to represent directionality was part of how the characters「上」and「下」came into being. Both originally used a long horizontal line as a reference, and a smaller mark representing the direction from the reference; the vertical strokes weren't added until later.

商
甲

前7.32.4「上」
合集102商
甲

甲942「下」
合集8943

---

The structure of「曰」is very similar to the structure of the synonym「言」.「言」was originally「舌」(tongue) with an additional horizontal mark on top.

商
甲

甲499
合集30697秦
簡

睡ㆍ秦1
　西漢
篆

老子ㆍ甲
　現代
楷


　

For reference,「舌」was originally a picture of a tongue sticking out of a mouth「口」with saliva marks.

商
甲


合集6248
篆

說文解字
　現代
楷


　

There is a further character which has the potential to cause confusion:「甘」(delicious > sweet) was also created from「口」, being originally a depiction of something tasty (represented by a mark) inside of「口」.

商
甲

乙1010
合集21731
篆

說文解字
　現代
楷

標楷體
　

---

How do you really differentiate them except by context?

Unfortunately, the mark in「曰」sometimes shifted position or was drawn right across the horizontal boundaries of the character. The regular script samples above are from a Taiwanese (標楷體) and Japanese font (EPSON正楷書体), respectively, and in the Taiwanese font the middle horizontal lines span both「曰」and「日」, while in the Japanese font the line does not span either character, making this at best an unreliable way of distinguishing the two.

As individual characters,「曰」and「日」should be distinguished by their width;「曰」is never taller than it is wide, while「日」is always taller than it is wide.

Even this method is unreliable when it comes to component parts of characters, because modern parts of characters obey proportion rules for aesthetic purposes rather than for semantics, and the shape of the component is no longer relevant in this case. This is most obviously demonstrated in the character「晶」, which is really three smaller「日」 stretched to various proportions in different fonts. Fortunately,

• 「日」is found as part of characters to do with time/brightness/rising/weather. If it's not on the left or right side, its position in a character may also provide a semantic hint, e.g.
  • 昇 (rise up)
  • 間 (gap > space, depicting sunlight streaming through a gap in the door)
  • 昏, 暮 (dusk)

• 「曰」and「甘」are both semantically connected to mouth and/or modified from the shape of「口」.

  • 會 (to meet, originally contained「口」modified to「曰」)
  • 旨 (delicious, originally contained「口」modified to「甘」, now looking like「曰」)
  • 曹, 者, 香 (contains distinguishing marks)

• How about when both semantic parts are found in one character?

  • 「昌」was the original character of「唱」, which was a compound indicating a morning _{(represented by a risen sun「日」)} call _{(originally「口」, now「曰」)} to work, extended to mean chant, song.

As components, we can only distinguish between「曰」and「日」based on the entire character, and not the shape of the component.

---

References:

• 季旭昇《說文新證》
• Multi-function Chinese Character Database
• Chinese Linguipedia
• 小學堂

Answer 2

In your question, you seem to already be aware of what differentiates 曰{いわ}く from 日{ひ}, which is that the middle stroke does not go all the way across, so that it does not touch the vertical stroke on the right side.

So I think the issue your having is the same as one that I remember having a long time ago when first learning kanji, which was an underappreciation of how exacting the Japanese language is about how kanji are written. For me, it was the difference between 土 and 士 that first confused and then educated me about how what I thought were subtle differences were actually big differences. In English, we can be a lot more flexible about how to write a character without losing the understanding of which character it is. Probably because with so many fewer characters, the risk of ambiguity is less, though that's just a pet hypothesis of mine.

In any case, for the most part, in Japanese small differences like that which differentiates 日{ひ} and 曰{いわ}く really do matter, and signify entirely different characters. (Which negates the need to answer your other question about how it is the same character is used for such different things, since they are not, in fact, the same character).

Take a look at this page, or this page, each of which provudes lists of kanji characters that look very similar. Note, for example, the subtle differences between 己{おのれ}, 已{のみ}, and 巳{み}.

On my computer, the difference between 日{ひ} and 曰{いわ}く is not clear because of the font I'm using. If that's the case with you, look at this page which shows a clear graphic of how to draw 曰{いわ}く, so you can see very clearly that the middle stroke is intentionally not touching the right side.

Having said all that, you should also be aware that sometimes there are small differences in the conventions of how kanji are drawn, so that there are cases where the same kanji can be drawn differently, depending on the font or the person writing it. Take a look at this question especially, and also maybe this question to explore those issues.

Hope that helps.