How to write the 匕 in 倾 ?

[Guest realmayo]

As I understand it:

When 匕 is written on its own, it has a 丿

But when it (or the component very like it) appears in e.g. 倾, it's written with a  一.

 

Is my understanding correct?

Is there anything (broader rule or context) to explain this?

Is it okay to page @Hofmann ?

[Tomsima]

Love questions like this!

 

When written on its own 匕 is supposedly written with a 丿in 楷書, but will still appear in some 隸書 with a 一. Outliers research says the earliest sources of 匕 tell us the character is a mixture of both a depiction of a spoon and a separate character, a flipped 人. This means that the 一/丿issue has for many centuries been ambiguous: essentially either way of writing can be considered correct, as the character itself isn't sure, as it were.

 

In the font I see on my computer right now the 匕 uses a 一, but if i want to input it with stroke-based input method I must select a 丿. If I select a different font, the very same unicode character will occasionally show the stroke as a 丿. So as you can see, this ambiguity still exists even in modernised forms.

 

If you look at more works by famous calligraphers, you will see that this character and many others are written in a variety of different ways. According to mood, ease of writing, and even whats aesthetically pleasing without being different enough to cause the character to be misunderstood, many calligraphers frequently made slight alterations to many many characters. A character such as 匕, which is already semantically ambiguous, could potentially be written in three different ways (一丿丶) in ONE text by the same callagrapher. The important thing to note is, for many people, this small variation simply didnt matter. This is true for your question on 倾 - either stroke is fine, as both are acceptable.

 

This perhaps isnt too helpful, as you will find the majority of fonts and written forms of 倾 will use 一. Some will use 丿, some will even use 丶, but 一 is the most popular, perhaps because it is easy to write, is clear, and has been used by influential calligraphers such as 顔真卿. The important thing to note is, before computers, before the printing press, there was no universal standard for every stroke of every character, only accepted norms, recognised forms that people wrote in according to region or in imitation of calligraphers. For example, a 王羲之 '云’ might suggest a 云 where the first stroke is a 丶 not a 一, the 'standard' is thus set by a person, or group of people, or a trend in a time or place. Not until computers came along did everything get set in stone as it were. That being said, as we saw above with 匕, even in computing there are many different appearances of the same character across different font styles, and different regions.

 

TLDR:

your understanding is basically correct

there is no rule, no standard, just trends spanning hundreds if not thousands of years

and yeah @Hofmanns book is great, get it.

[889]

"A character such as 匕, which is already semantically ambiguous, could potentially be written in three different ways (一丿丶) in ONE text by the same callagrapher."

 

I thought there was a tradition in Chinese calligraphy that an artist would purposely NOT write a repeated character the same way: he'd use a variant both to show his skill and keep the work as a whole varied and interesting.

[Hofmann]

The grapheme 匕 can be executed in a few ways. The first stroke can be 一 or 丿. 一 was most popular until Ming or Qing, except in 化 and 死, in which 匕 appears as it does here. The second stroke can be, in order of appearance, ㇗, ㇄, or ㇟. When 匕 occurs left of something, it most commonly features ㇗ (the 一 part replaced by ㇀, a common left-position transformation) and sometimes ㇄. Rarely ㇟ as that requires breaking a rule.

Note there is no reliable differentiator between the two homographs written with the grapheme 匕.

Also, there is no tradition of deliberate variation in handwriting. Stuff just comes out however it comes out. It varies because humans are imprecise and varying levels of lazy.

[Tomsima]

The idea of not repeating a character in the same way does exist, but it relates to using different angles, stroke thickness and varying degrees of 行/草 styles of writing to draw out the variation of a character that appears a lot in one text, not to using variant forms of characters (unless you want to go down the 壽/福 route...). This is easy to spot if you look at calligraphers copies of 道德經, where there are many, many 之、得、道、天、人 etc. that must be written out but have no variant form.

 

I suppose you could make an argument that 顔真卿 might have used a lot of variants for the reason of keeping things interesting, as his and similar styles of 楷書 are (by definition) 'regular' in style...

[889]

This is an example, quickly found on Google, of the sort of thing I've read that led me to believe differently:

 

"在书法作品中，如果出现同一个字，或者在一个字里出现同样的笔划，都必须变换写法，这已经是书法创作约定俗成的规律和要求。写异体字正好可以满足这个要求 。 "

 

http://www.sohu.com/a/124602356_429316

[Hofmann]

The variation I'm referring to includes both variants (primarily from laziness) and stroke shape and placement (from imprecision). I don't see enough variation in 顏真卿's regular script to conclude that he deliberately added variation.
 

889, I was aware of that sort of thing. It's a misconception, from mistaking natural variation for deliberate variation.

 

Oh, @realmayo, standard characters, such as Simplified Chinese, are quite specific about how graphemes and components are written. If you're following a standard, then it must be as prescribed, in the case of 倾, 匕 must feature 一.

[Tomsima]

Im guessing your search terms led you to what you wanted to find. This article comes from a 公衆號 called 書畫少年俱樂部, its short and sweet - trying to encourage beginners to not be put off when they start learning calligraphy and encounter not only 繁體字 but also many 異體字 and small variations in characters.

[Hofmann]

Good thing I don't practice or teach calligraphy.

[889]

I only know what I read, and I'd read that same assertion before elsewhere. Right or wrong, it seems a common enough belief.

 

And I did find a source for the assertion.

[Guest realmayo]

Thanks all.

 

18 minutes ago, Hofmann said:

If you're following a standard, then it must be as prescribed, in the case of 倾, 匕 must feature 一.

 

Do you know whether that presription forms part of a larger rule and if so how it usually applies and what its motivation is? I vaguely remember you (I think) complaining (I think) that the standard 起 is bad because the final stroke of both the 走 and the 己 are written out in full, whereas typically only one of them should be.

Is there, for instance, a convention that the stroke in question should be 一 unless the 匕 features on the right, exposed side of the character (能) etc..? And is there some broader aesthetic or mechanical reason to limit the right-exposed strokes to just one full one?

[edelweis]

hm never mind.

Edited April 3, 2018 at 09:05 PM by edelweis

[Hofmann]

On 4/3/2018 at 10:20 AM, realmayo said:

Is there, for instance, a convention that the stroke in question should be 一 unless the 匕 features on the right

That came first, and did not become 丿 because there was less pressure to evolve to it. The tendency to write 匕 with 丿 and 乚 on the right is from a desire for uniformity (and therefore simplicity) in the script.

 

On 4/3/2018 at 10:20 AM, realmayo said:

is there some broader aesthetic or mechanical reason to limit the right-exposed strokes

That's a feature carried over from clerical script and cursive (行書). I don't know why it existed in clerical script.

 

On 4/3/2018 at 10:13 AM, 889 said:

Right or wrong, it seems a common enough belief.

You probably know this, but that many people believe something doesn't affect your p.

 

All valid linguistics is descriptive. A valid description of a script is based entirely on observable script features. Regarding the feature in question, deliberate variation of the glyphs of the same grapheme, I am failing to reject the null hypothesis that there is no such feature because I would rather make a Type I error than a Type II error. My conclusion is supported by a few things:

• If such a "meta" feature existed in Chinese, it would likely also exist in other scripts, yet I haven't seen anything like it. Also, Chinese is not special.
• The examples shown that are claimed to exhibit the feature don't have more variation than examples that are not.
• It is not humanly impossible to write a grapheme the same way twice. To aim to do something that is impossible not to do is an unlikely expenditure of effort.

An experiment would be easy enough: Ask a bunch of people to write a sentence with more than one instance of a few graphemes. Then ask them if they meant to write the different looking ones differently. If you can reject the null hypothesis with a p value less than 0.05, you'll rock the world of linguistics.

[dwq]

On 4/4/2018 at 12:20 AM, realmayo said:

Do you know whether that presription forms part of a larger rule and if so how it usually applies and what its motivation is?

 

I'm interested in knowing how the standards chose the "correct" forms as well.  It almost appears to me that sample characters (which contains errors and inconsistencies) are included as examples and then somebody took the samples too seriously.  Are they really rigorously selected?

 

OTOH, by claiming one standard you can cut down on questions like "what is the correct way of writing this character?".  The problem is that when you have multiple standards (Mainland, Taiwan, and Japanese, etc.) you get answers like "You write it this way, but only in Taiwan".  Also, now you get questions like "why is the way this component is written differently when it is part of this other character, in this standard?"

 

I find it amusing whenever a computer font gets updated and a glyph changes form to "conform to a national standard", and people complain now they cannot write it the way they liked, or documents look "different" or "incorrect" when viewed on a different version of an OS.  (I totally sympathize with them, it is not an elegant solution to the problem)

 

[Hofmann]

11 hours ago, dwq said:

Are they really rigorously selected?

No. They can be quite arbitrary and/or subjective. For example, Taiwan's reason for keeping the hook on 起 was "為求美觀." (Here's the whole document. Here's Hong Kong's counterpart.) 

[edelweis]

23 hours ago, Hofmann said:

Ask a bunch of people to write a sentence with more than one instance of a few graphemes.

Tell a bunch of random people "please handwrite this sentence"

=/=

Tell a bunch of calligraphy enthusiasts "please create a calligraphy with this sentence"

 

[Hofmann]

Great. What sort of people are we talking about here, what's a calligraphy enthusiast, and what's calligraphy?