What 𗍵 means
It is a historical Tangut character classified as a transliteration.
Tangut was the language this medieval script was created to write. Li Fanwen’s dictionary indexes this entry as “a transliteration”; that label describes its role, not a translation into English.
Its reading is reconstructed, or estimated from historical evidence: Li Fanwen gives *tśiwa¹, Miyake gives *chwa₂, and Kychanov and Sofronov give *tśwâ¹. These are specialist spellings of estimated sounds, not ordinary words to sound out using English letter names.
Where does 𗍵 come from?
It appears in 12th-century Tangut dictionaries.
An early record is in the Wénhǎi (“Sea of Writing”), at entry 26.152; the Tóngyīn (“Homophones”) also records it. These medieval references show that the character predates digital encoding.
Tangut writing was created around 1036 under the first Western Xia emperor to write the Tangut language. After Western Xia fell, it continued in official, private, and religious texts before becoming obsolete.
Historical and scholarly; it belongs in discussion of Tangut writing, not ordinary conversation.
Don’t call it Chinese: Tangut writing resembles Chinese ideographs, but the scripts are distinct.
Who writes 𗍵
Loading…
Questions people ask
3Is 𗍵 a Chinese character?
No. It is Tangut. Tangut writing resembles Chinese ideographs, but it is a distinct script and cannot be read as Chinese.
Why does 𗍵 show as a box?
A font or app without Tangut coverage can display a blank or replacement box. The underlying text may still contain 𗍵.
Can I type 𗍵 with a normal keyboard?
Usually not directly. An ordinary keyboard has no dedicated key for this historical character; use Unicode code-point entry or a Tangut-specific input method.
How to type or find 𗍵
- Search it as
- Tangut character · Tangut transliteration · Tangut 17375 · 西夏文 · tśiwa · chwa · tśwâ · Wénhǎi
- Shows up in
- Anything that takes text · its shape is whatever your font draws
𗍵\17375Codes and encoding1 code points
A visible symbol can contain several code points. JS .length counts UTF-16 units; UTF-8 counts bytes.


