莞 (guǎn): used in 东莞 (Dongguan, a city in Guangdong), Skimmia japonica

莞 (guǎn) is a Chinese character meaning “used in 东莞 (dongguan, a city in guangdong).” Outside the HSK 3.0 syllabus (CLEC 2022), it is composed of 艹 (semantic) and 完 (phonetic). It ranks #5193 in character frequency (SUBTLEX-CH corpus).

Etymologically derived, 莞 plant. Its radical form 艹 (grass) appears in many related characters such as 蒸 (zhēng, to steam), 草 (cǎo, grass), 花 (huā, flower).

Spoken pronunciation

Definitions

  1. used in 东莞 (Dongguan, a city in Guangdong)
  2. Skimmia japonica

Etymology & Origin

莞

pictophoneticplant

Decomposition: ⿱艹完 (layout: top-bottom)

Stroke Order

1
2
3
4
5
6
7
8
9
10

Practice writing 莞 with real-time feedback — trace each stroke in the correct order and build muscle memory in the HanziFeed app.

Words & Compounds

Common Compounds

WordPinyinMeaning
东莞proper-noundōng guǎnsee 東莞市|东莞市
莞尔一笑wǎn ěr yī xiào(literary) to smile
东莞市dōng guǎn shìDongguan, prefecture-level city in Guangdong Province 廣東省|广东省
莞尔adjectivewǎn ěr(literary) to smile
4
Total compounds
50
As first character
25
As last character
25
As middle character

莞 appears in 4 compound words: 50 as the first character, 25 as the last, and 25 in a middle position. Compound statistics computed from SUBTLEX-CH and HSK 3.0 vocabulary data.

Strongest Collocations

Characters that most frequently co-occur with 莞 in natural Chinese text, ranked by NPMI (Normalized Pointwise Mutual Information) — a statistical measure of association strength.

东dōng
0.42720 co-occurrences

Example Sentences

AI-Generated

这个莞字在古汉语中有不同的含义。

zhè gè guān zì zài gǔ hàn yǔ zhōng yǒu bù tóng de hán yì .

In Classical Chinese, the character "莞" has different meanings.

中国新闻网·文化Sep 2026

因此,他只能同情地看待,“莞尔一笑,不痛不痒”。

yīn cǐ , tā zhī néng tóng qíng dì kàn dài , guān ěr yī xiào , bù tòng bù yǎng .

Therefore, he could only view the situation with sympathy, “smiling faintly, without any real emotion.”

中国新闻网·文化Aug 2026

电影里,当“永乐”两字反复出现时,也许熟悉八仙故事的观众会莞尔一笑。

diàn yǐng lǐ , dāng yǒng lè liǎng zì fǎn fù chū xiàn shí , yě xǔ shú xī bā xiān gù shì de guàn zhòng huì guān ěr yī xiào .

In the movie, whenever the words “Yongle” appear repeatedly, viewers familiar with the story of the Eight Immortals may find themselves smiling.

Character Family

Semantic Family — Characters sharing the 艹 (plant) component

Frequently Asked Questions

What does 莞 (guǎn) mean in Chinese?
莞 (guǎn) primarily means "used in 东莞 (dongguan, a city in guangdong)." It is not part of the HSK 3.0 syllabus. It ranks #5193 in character frequency.
What's the difference between 莞 and 院?
莞 (guǎn) and 院 (yuàn) are often confused. confusable. The key distinguishing feature: 艹 vs 阝 (same 完 component).
How many strokes does 莞 have?
莞 is written with 10 strokes. The correct stroke order matters for recognition and handwriting speed — practice with the animated guide above to build proper technique.
What is the radical of 莞?
The radical associated with 莞 is 艹 (grass). This radical appears in many characters related to grass.
What are the components of 莞?
莞 is composed of: 艹 (semantic), 完 (phonetic). Its IDS decomposition is ⿱艹完 with a top-bottom layout. Understanding the components helps with both memorization and recognizing related characters.
What are common words containing 莞?
Common words with 莞 include: 东莞 (dōng guǎn, "see 東莞市|东莞市"); 莞尔一笑 (wǎn ěr yī xiào, "(literary) to smile"); 东莞市 (dōng guǎn shì, "Dongguan, prefecture-level city in Guangdong Province 廣東省|广东省"); 莞尔 (wǎn ěr, "(literary) to smile"). There are over 4 compound words containing this character.
Is 莞 the same in simplified and traditional Chinese?
Yes, 莞 is written the same way in both simplified and traditional Chinese.

Practice writing 莞 with real-time feedback

Trace stroke sequences, hear spoken pronunciation, and build lasting retention with spaced repetition in the HanziFeed app.

Character data sourced from Unihan (Unicode Consortium), SUBTLEX-CH frequency corpus (Cai & Brysbaert, 2010), and Make Me a Hanzi (stroke data). Collocation strength measured via NPMI (Normalized Pointwise Mutual Information). Verified by the HanziFeed linguistics team.

HSK classification follows the HSK 3.0 Standard (Center for Language Education and Cooperation, CLEC, 2022 revision). Idiom data from Wiktionary (CC BY-SA 3.0).

Data last verified: March 2026.