unify to ㇀ (U+31C0)?

The current script for U+31C0 ㇀ is common (Zyyy), but it also has the Script_Extensions value as Han (Hani).

We generally don't believe a standalone Han character can be unified with a stroke by design.

Additional information:

《説文解字》「孑：無右臂也。从了，乚象形。」
Thus we believe that character is just a reflection of 乚 based on the regular script shape, and not a mere stroke.

The quoted part from 說文 (ShuowenJiezi) in post-Sung edition for "孑" is like these:
(南宋刊元修大徐本(海源閣本), 巻14下 葉6左)

(清道光19年 祁寯藻刊 小徐本, 巻28 葉13左)


The "L" like glyph in SW description is not "𠃊" (of 直) or "乚" (of "礼"), as the description says as 象形, and no self-standing entry is found in SW. In my understanding, "(了 is a figure without both arms, and) 孑 is a figure without right arm". I guess, the glyph in 慧琳一切経音義 might be the explanation influenced by 楷書 shape of "孑", and it is difficult to conclude as "㇀"-like glyph as self-standing character, meaning "left arm", by 慧琳一切経音義.

"無右臂從了象形㇀聲也㇀音厥" in 慧琳一切経音義 is slightly hard for me to understand precisely. "㇀聲也㇀音厥" looks as if this character were 形聲 composition (radical & pronunciation), but if so, why "從了象形"? In post-Sung dynasty versions of 説文, 形聲 composition is regularly described as "從A,B聲". In the 説文 quoted in 慧琳一切経音義, "従A象形B聲也" is frequent?

After the punctuation mark is added to the 一切經音義, the sentences are "說文，無右臂。從了，象形。㇀ ，聲也。㇀音厥". 一切經音義 did not specify that this(㇀) was a character, just a marker, indicating its reading.
In 說文, "(孑)，無右臂也。从了、乚(𠃊)，象形"
說文(大徐本and段注) is composed of "了" and "乚(𠃊)", which is a pictograph. 乚(𠃊) indicates a left hand of (depending on) a body (了). Without the body, it means nothing. So 乚(𠃊) shall not be a character.
The "六書" should be "omitted from the right hand(省去右臂)" of “Hieroglyphs with omitted strokes(省體象形)”.

See a current surviving edition 小徐本 (說文解字繫傳) of 說文解字:
https://rmda.kulib.kyoto-u.ac.jp/item/rb00031845




We see an annotation by 徐鍇 that he believes 孑戉 contain the questioned right hook character, which corresponds to the Regular script 𠄌 (U+2010C), so that we believe there was an established recognition (at least) in the Song period that 孑 uses 𠄌 as a component.

Some may have suspicion that how a character could have phonetic components when it belongs to 象形, but it is not uncommon even in the current versions of 說文解字 to show alternative analyses, as in the description of 孒 in the image above, or following examples:

主：鐙中火主也。从■（⿱凵土），象形。从丶，丶亦聲。
履：足所依也。从尸从彳从夊，舟象履形。一曰尸聲。

The pronunciation given for 𠄌 in 說文解字繫傳 is 倶越反 = 厥.

Thus we believe that there is no contradiction and many circumstantial supports to think the one-straight-stroke glyph in our evidence is meant to be a variant of 𠄌, only deformed to match the Regular script shape of 孑, either because 慧琳 saw fit, or the version of text he quoted already looked like that.

Regarding the comment #9364, the suggested punctuation:

「說文，無右臂。從了，象形。㇀ ，聲也。㇀音厥。」

is not likely considering the convention (體例) seen in quotations of 說文解字 by 慧琳音義. Recent critical edition 《一切經音義三種校本合刊》 also reads as:

「說文：無右臂。從了，〔乚〕象形。㇀ 聲也。㇀音厥。」 (p. 1582)

which means that this proposed character is the phonetic component of 孑. Generally, it is not prohibited that X聲 directly follows the 象形 declaration in 慧琳音義. It could be 慧琳's preference or a common style shared by texts at that time. Examples include:

「説文鐙中火主也象形從丶聲」（卷三十一）
cf. 大徐本「主：鐙中火主也。从■（⿱凵土），象形。从丶，丶亦聲。」

「説文從尸彳久[夊]舟象形尸聲」（卷三十二）
cf. 大徐本「履：足所依也。从尸从彳从夊，舟象履形。一曰尸聲。」

「盾所以扞身蔽目也以自蔽從十目象形厂聲」（卷六十七）
cf. 大徐本「盾：瞂也。所以扞身蔽目。象形。」
cf. 《說文解字注》「从目。（各本少二字。今依玄應補。）象形。（鍇曰：𣂑象盾形。按今鍇本或妄增厂聲二字。）」

Especially the 說文解字注 comment in the last example, contrary to the speculation that 「厂聲」 in 小徐本 might have been 徐鍇's invention, suggests the possible existence of such convention that predates the Song dynasty.

I am still not in favor of encoding this character as a CJK Unified Ideograph. The strokes that are already encoded, about half of which are in Extension B, do not represent a valid precedent for doing so. There are 36 strokes in the CJK Strokes block, but there are only 22 CJK Unified Ideographs that have a kTotalStrokes property value of 1, which are the only ones that could be treated as strokes. Also note that we have encoded Japanese kana digraphs and even bopomofo as CJK Unified Ideographs, and I think that we should treat strokes the same way. In other words, treat them as mistakes, and make sure that we do not encode any more as CJK Unified Ideographs. I would also like to point out that any character can have a reading (or name), including the strokes in the CJK Strokes block. We simply cannot document them in the Unihan database. Appendix F of the Core Specification, "Documentation of CJK Strokes," provides names for the strokes.

I suspect that comment #9396 still wrongfully believe that we are "pushing something which is a stroke's character-ification into CJKUI". Indeed two kinds of arguments were made in support of encoding of this character in the Oct 19 discussion concurrently, which might have led to the confusion.

(1) This is a character which only has one stroke, thus is a CJKUI (maintained by me, etc.)
(2) Even if this is not strictly a character, it is useful to encode it for completeness and good evidence support (maintained by Henry, etc.)

I reiterate that our stance has been (1), and our point is already covered in the comment #9373. But let me recap the whole reasoning with full English translation for those who are not familiar with Middle Chinese.

a) text in our Evidence 1 & 2:
……說文：無右臂。從了，象形。㇀ 聲也。㇀音厥。
"Shuowen Jiezi: 'having no right arm. Pictographic character based on 了. Sound of ㇀.' ㇀ reads as 厥."

b) entries in 說文解字繫傳, vol. 24 (1st picture of comment #9373)
亅：鉤逆者謂之亅。象形。凡亅之屬皆從亅。讀若橜。臣鍇曰：鉤喙之曲𦬆[芒]。今曰逆須。孒[in seal script]從此。瞿月反。
"亅: barb of hook is called 亅. Pictographic. All characters under the radical 亅 are based on 亅. Read as 橜. Xu Kai's note: the bent tip at the end of a hook. Today we call it 逆須 (turned beard). 孒 is based on the character. Alliterates with 瞿, rhymes with 月."

𠄌：鉤識也。從反亅。讀若罬。臣鍇曰：鉤柄之表識。孑[in seal script]、戉從此。俱越反。
"𠄌: hook mark. Based on reversed 亅. Read as 罬. Xu Kai's note: symbol looking like a hook's stem (= checkmark). 孑 and 戉 are based on the character. Alliterates with 俱, rhymes with 越."

c) entries in 說文解字繫傳, vol. 28 (2nd picture of comment #9373)
孑：無右臂也。从了，㇄。象形。經節反。
"孑: having no right arm. Based on 了 and ㇄. Pictographic. Alliterates with 經, rhymes with 節."

孒：無左臂。象形。从了，㇓聲。俱越反。
"孒: having no left arm. Pictographic. Based on 了 and sound of ㇓. Alliterates with 俱, rhymes with 越."

With a) and c), we know the ㇀ -shaped glyph is meant to be the ㇄-shaped glyph in 說文解字繫傳 (which is effectively 小徐本 of 說文解字), but since 說文解字 list all head-characters in Seal script while description in Regular script, we still don't know what the identity of this glyph is. However, b) proves that this character is actually 𠄌 in the same book, which is a character that represents a word "hook mark".

The pronunciation of the proposed character is given as 厥 according to our evidence, which is in MC notation 臻合三入月見, or reconstructed by Wang Li as *kĭwɐt. The pronunciation of 𠄌 according to 說文解字繫傳 is 俱越反, which also indicates 臻合三入月見, or Wang Li *kĭwɐt. This perfectly supports that our ㇀ is a variant of 𠄌.

-----

Comment #9396 further states: " I would also like to point out that any character can have a reading (or name), including the strokes in the CJK Strokes block. We simply cannot document them in the Unihan database. Appendix F of the Core Specification, "Documentation of CJK Strokes," provides names for the strokes."

According to Appendix F, the stroke U+31C0 ㇀ has a name 提 (tí). Pronunciation of this word in MC is 止開三平支常, or Wang Li *ʑǐe, which is totally different from what is discussed here (not to mention U+31D9 ㇙ a.k.a. 竪提). This is parallel to the situation that a single horizontal line, U+31D0 ㇐, has a name 橫 (MC 梗合二平庚匣; Wang Li *ɣwɐŋ) but a character that only consists of this stroke exists as U+4E00 一 (MC 臻開三入質(A)影; Wang Li *ǐět) and means "one". Unless the comment intends that U+4E00 一 should have not been encoded because "yī" is merely a "name" of this stroke, we don't think the logic applies to this case.

Yes, I understand all of that. My point is about this character's actual real-world usage after it has been encoded as a CJK Unified Ideograph, which will be as a clone of U+31C0 ㇀ CJK STROKE T. If the consensus of the IRG is to encode this character as a CJK Unified Ideograph, then I will accept that.

Apologies for invoking the Tenth Man Principle, but it is necessary for this particular case.