Hanzi Mistakeholders
Lazily borrowing everything from Language Log, I offer up “mistakeholders” for contemplation:
The people who have come to rely on features that are actually implementation errors are called ‘mistakeholders’.
My assertion: there has GOT to be an example of a Chinese character created in error through mis-copying, mis-remembering, what-have-you, that subsequently was accepted into the realm of legitimate hanzi. The existence of such would make us (users of the script) all mistakeholders.
Sounds like a job for a hound of hanzi history. Anyone?
There are a number of so-called vulgar variants such as 靣 for mian and 囬 for hui which lasted long enougj to get encoded. Variants in general may account for a great deal of mistakeholders asuming at lesst some aren’t the “origonal” form.
Others would argue anyone using jianti is a mistake holder, but thst’s neither here nor there.
This may be an example of what you are looking for. 得 dé ‘to get’ is originally composed of three elements, as revealed in early textual sources: 彳 on the left, 貝（贝）on the top right, and 寸 on the bottom right. The two elements on the right depict a hand (寸) grasping something valuable (貝-in early form, written without the two “legs” at the bottom). Over time, the lowest horizontal stroke got detached from 貝, leaving behind just 日, and it was reinterpreted as being on top of 寸. If you think about, the bottom right part of 得 is pretty weird — as far as I know it doesn’t occur in any other character. Once you realize the original form of the character, it all makes more sense.
Zev — the 得 example was exactly what I was thinking about. Very cool, esp since I specifically remember writing 得 wrong a number of times before I finally learned to put the extra heng on top of the 寸. Assuming you’re right that that particular component doesn’t appear anywhere else, I feel more justified in mucking it up!
Interesting stuff. Though I take issue on the “that components doesnt appear…” part, if only on the grounds of the current form’s components being 彳, 旦 and 寸.
This is of course only from a visual perspective and not one taking into account the historical form of the character. In other words, I’m saying this knowing that it’s a mistake and I’m therefore wrong.
In a sense you are correct, but I would guess that most native speakers of Chinese don’t think of the right side of the character as made up of 旦 and 寸, but rather of 日 and the part below it. I can’t type that part (寸 with a horizontal stroke atop it) into this comment because it’s not in any font. I’d be curious to hear from some native speakers about how they conceptualize the components of 得.
Asking around, that is, using the sample of 3 that are my flatmates, they’re all with you, Zev. The reason given is that the 一 and the 寸 are touching. In the Sinoglot tradition of getting sample sizes that are way too small, I’m going to call it good.
How about 碍?
http://www.nciku.com/search/zh/detail/%E7%A2%8D/137
石，旦，寸。
Don’t know what to tell ya. Seems right to me.
Kellen, I think Aaron’s responding to Zev, who didn’t know of any other char that had 寸 with a heng on top. But there it is. And wouldn’t you want to avoid the same 旦 + 寸 component analysis that you just disavowed? 😀
Of course, maybe, like usual, I’m missing something
My only point was that 得 is not the only character with that particular composition on the right-hand side. In fact there’s also 棏 and 淂.
Yeah. Sorry Aaron.
The breakdown I gave is actually right off the nciku page to which I linked. I don’t know if other apps/systems would give the same. Dimsum stopped working on my computer after I upgraded the OS so I can’t check that.
Personally I don’t think it ought to be avoided. While my resident Mandarin speakers seem to think 一寸 is one chunk, they’ve been wrong before.
Maybe a better search than the one for 旦寸 as the righthand component would be another instance of 一寸 in another character, sans 日?
You are absolutely right, I wasn’t correct in saying that the thing that looks like 一 over 寸 occurs only in the character 得. As you pointed out, it also occurs in other characters having the same right side: 碍, 棏, 淂 and so on. But this amounts to the same thing: 一 over 寸 only occurs because of “mistaken” reanalysis of 目 (a variant of 貝) over 寸. I still maintain that 一 over 寸 is not in and of itself a meaningful recurring character component, and I think you won’t find it occurring as such.
There’s nothing wrong with giving a mechanical breakdown of the modern form of the right-side of these characters as 旦 over 寸. From a purely graphic viewpoint that is self-evidently accurate. But that is not indicative of the historical origin, and I think it’s likely (as Kellen’s flatmates seem to concur) that that is not how native users of the script conceive of the structure.