Logos52
wiki / Language / Chinese / Form Components

Form Components

Concept updated 2026-09-11

Form Components

A form component is a part of a character that helps the character’s meaning by depicting something, and it keeps that job after the modern strokes have stopped looking like the thing depicted. A meaning component helps the character’s meaning by what the component itself means. Form components and meaning components are the two subtypes of a semantic component, which is any part that contributes to the meaning of the character it sits in. Meaning Components covers the other subtype.

“Depicts” is a statement about where the component came from. It says nothing about whether the printed shape still looks like the object today.

Form components are one topic under the hub How Chinese Characters Work. Surface vs Deep Structure covers how far to divide a single character into parts.

Core takeaways

  • A form component helps its host character by a picture of something, and it keeps that job after the modern strokes stop looking like the thing.
  • When a character is built from more than one form component, the scene the parts make together is the meaning. 取 is a hand 又 seizing an ear 耳.
  • The modern shape can mislead. The top of 美 is a feather headdress on a standing adult. Reading it as 羊 is a mistake.
  • Look at the ancient form once, at first encounter. That one look gives the scene, and a wrong guess from the modern shape never takes its place.
  • A character used as a word and the same shape used as a component have different lists of meanings. 心 as a word means mind and center; as a component it means thought and feeling.
  • Loan senses such as 然 “like this” sit outside the picture. Do not derive them from it.

Characters that began as drawings

Early characters drew the things they named: 犬 drew a dog, 山 drew peaks, 耳 drew an ear, 网 drew a mesh. Those drawings are the form attribute described in Three Attributes of a Character, the written half of a spoken word.

The usual link between a semantic part and the meaning of the character is loose. The link is tight when the character is built from more than one form component, because in that case the scene the parts make together is the meaning.

Characters built from more than one picture

取 is 又, a hand, seizing 耳, an ear. Armies counted casualties in ears. The senses extend to take, receive, and select. The ear count is the scene, and the later senses grew from it.

美 is 大, an adult seen from the front, wearing a feather headdress. The modern top has decayed into the empty component 𦍌; Empty Components covers that leftover headdress. Reading the top as 羊 misreads the scene. The sheep reading is a common wrong parse. The top is still debated: headdress, goat horns, or “big sheep”. Of those three, the headdress is the reading taken as correct.

羅 is net 罒 plus bird 隹, with 糸 added later to stress threading. A character can gain components over time to make a meaning more exact, so a character is not always one drawing made at one time.

The picture points at the original meaning

然 originally meant “to burn”: 灬 depicts fire, and 肰 gives the sound. 狗 is 犭 plus the sound 句. The form component points at the original meaning of the character. It does not point at a loan sense that the character took on later.

光 depicts a person holding a torch. The person was originally a kneeling figure 卩 and is now written 儿. The corrupted top no longer looks like 火. The scene stays the correct parse after the resemblance is gone. 光 is a case where the definition of a form component can be checked.

大 depicts a standing adult and means “big.” As a component it almost always depicts a person. Only a few characters use its meaning, and 尖 is the named exception. Three Attributes of a Character covers all three jobs of 大. For form components, the person reading and the one meaning exception are what matter.

木 depicts a tree, and “tree” survives among its senses, so depiction and meaning coincide. In that case, classify the component as semantic and go no further. That is the murky-case rule, and it stops over-classification.

Looking up the ancient form once

The ancient form turns the stroke pattern back into a scene that can be pictured. The memorization rules Visualization and Association work on a scene, and the rule Understanding is met once the scene is known. Rules of Effective Memorization covers those three named rules.

Look at the older drawing once, at first encounter. Without that look, a folk parse of the modern shape takes the place of the scene; 美 read as a big sheep is the type case. The habit is this: look the ancient form up once, then stop consulting it. After that one look, 美 is a person in a headdress, the modern strokes can be derived from the scene, and the sheep reading no longer fits anything.

Once the scene is known, The IME Method rebuilds a forgotten form from candidates.

A word and its component form keep two lists of meanings

A character used as a word and the same shape used as a component keep two lists of meanings. 心 as a word covers mind and center; as a component it covers thought and feeling. 頁 inside other characters keeps only “head.” 山 widens to rock, 貝 widens to money, and 犭 widens from dog to animals generally. Meaning Trees and Original Meanings covers the meaning trees. The two lists are not the same list.

Extension chains start at the picture. 刀 runs from knife, through knife-shaped bronze currency, to the modern PRC slang for “US dollar”. That last sense is slang and is not part of the character’s core. 車 runs from war chariot, through cart and train, to vehicles in general.

Loan senses sit outside the scene. Such senses are marked 〇: 然 “like this,” 足 “sufficient,” 頁 “page.” Deriving those senses from the picture produces a false etymology. Meaning Trees and Original Meanings covers the 〇 system.

One component, several printed shapes

Treat positional variants as one mental entry plus a rule about position: 犬 becomes 犭, 肉 becomes 月, 心 becomes 忄 or ⺗, 手 becomes 扌, 刀 becomes 刂. 月 standing for 肉 and 月 standing for the moon are the same printed shape in two different host characters. They are not two words.

网 changes shape by function: it is written 罓 when it is a sound component and 罒 when it is a semantic component. Whether any other component does this is an open question.

儿 as a component, including inside 兒, is a variant of 人 and reads rén. The native name is 儿字底. The standalone simplified character 儿 is 兒 with the top deleted. Reading 儿 as “legs” misparses a person. Three Attributes of a Character gives those two histories.

攵 was simplified from 攴, “to strike”, which drew a hand holding a whip. It is easily confused with 夂 and 夊, both of which are variants of 止 “foot.” 攴 and 攵 occur almost only inside other characters.

广 keeps its own identity in traditional script. Only simplified script uses it to write 廣. Traditional script is the target script.

虫 originally wrote 虺, a venomous snake. It has also written “insect” (蟲) since Qin. Simplification kept the first of the three 虫 in 蟲.

How to practice this

  1. When you meet a new character, look up its ancient form once. Notice whether the old drawing gives you a scene that the modern strokes do not.
  2. Parse the top of 美. Notice that reading it as 羊 gives no scene, and that a feather headdress on a standing adult does.
  3. Take a character with two form components, such as 取. Notice that the two pictures make one scene, and that the scene is the meaning.
  4. Take 心 and list its meanings as a word, then its meanings as a component. Notice that the two lists differ.
  5. Take a sense such as 然 “like this” and try to reach it from the picture of fire. Notice that no path exists, and mark the sense 〇.
  6. When you see 犭, 月, 忄, 扌, or 刂, name the full form: 犬, 肉, 心, 手, or 刀. Notice that one entry plus a position rule covers each pair.

Open questions

  • Which characters besides 尖 use 大 for its meaning rather than its picture.
  • Whether any component besides 网 changes its allograph, its printed shape, by function.
  • A lookup workflow for ancient forms that holds up at real reading volume. Without one, the habit of looking once at first encounter stops working at scale. None has been worked out.

Sources

  • Qiu Xigui. Chinese Writing. Early Translation Series, Society for the Study of Early China / Institute of East Asian Studies, UC Berkeley. Pictographs as origin; depiction decays under stylization.
  • DeFrancis, J. (1984). The Chinese Language: Fact and Fantasy. University of Hawaii Press. “From Pictographs to What?”: resemblance is historical, not a property of the modern glyph.