Chinese classifier

The phrase 车辆 (chē-liàng, car-CL) has the classifier after the noun. It could refer, for example, to "the cars on the road".

In addition to their uses with numbers and demonstratives, classifiers have some other functions. A classifier placed after a noun expresses a plural or indefinite quantity of it. For example, 书本 (shū-běn, book-CL) means "the books" (e.g. on a shelf, or in a library), whereas the standard pre-nominal construction 一本书 (yī běn shū, one-CL book) means "one book".[8]

Many classifiers may be reduplicated to mean "every". For example, 个个人 (gè-ge rén, CL-CL person) signifies "every person".[note 3][9]

A classifier used along with 一 (yī, "one") and after a noun conveys a meaning close to "all of" or "the entire" or "a ___full of".[10] The sentence 天空一片云 (tiānkōng yī piàn yún, sky one-CL cloud), meaning "the sky was full of clouds", uses the classifier 片 (piàn, slice), which refers to the sky, not the clouds.[note 4]

Classifiers may also indicate possession. For example, the Mandarin equivalent of "my book" would often be 我的书 (wǒ de shū, me-'s book), but in Cantonese this would typically be expressed as 我本书 (ngo⁴ bun² syu¹, me-CL book), with the classifier serving as a possessive marker roughly equivalent to English 's.

Within the set of nominal classifiers, linguists generally draw a distinction between "count-classifiers" and "mass-classifiers". True count-classifiers[note 8] are used for naming or counting a single count noun,[15] and have no direct translation in English; for example, 一本书 (yì běn shū, one-CL book) can only be translated in English as "one book" or "a book".[20] Furthermore, count-classifiers cannot be used with mass nouns: just as an English speaker cannot ordinarily say *"five muds", a Chinese speaker cannot say *五个泥 (wǔ ge nì, five-CL mud). For such mass nouns, one must use mass-classifiers.[15][note 9]

Mass-classifiers (true measure words) do not pick out inherent properties of an individual noun like count-classifiers do; rather, they lump nouns into countable units. Thus, mass-classifiers can generally be used with multiple types of nouns; for example, while the mass-classifier 盒 (hé, box) can be used to count boxes of lightbulbs (一盒灯泡 yì hé dēngpào, "one box of lightbulbs") or of books (一盒教材 yì hé jiàocái, "one box of textbooks"), each of these nouns must use a different count-classifier when being counted by itself (一盏灯泡 yì zhǎn dēngpào "one lightbulb"; vs. 一本教材 yì běn jiàocái "one textbook"). While count-classifiers have no direct English translation, mass-classifiers often do: phrases with count-classifiers such as 一个人 (yí ge rén, one-CL person) can only be translated as "one person" or "a person", whereas those with mass-classifiers such as 一群人 (yì qún rén, one-crowd-person) can be translated as "a crowd of people". All languages, including English, have mass-classifiers, but count-classifiers are unique to certain "classifier languages", and are not a part of English grammar apart from a few exceptional cases such as head of livestock.[21]

Within the range of mass-classifiers, authors have proposed subdivisions based on the manner in which a mass-classifier organizes the noun into countable units. One of these is measurement units (also called "standard measures"),[22] which all languages must have in order to measure items; this category includes units such as kilometers, liters, or pounds[23] (see list). Like other classifiers, these can also stand without a noun; thus, for example, 磅 (bàng, pound) may appear as both 三磅肉 (sān bàng ròu, "three pounds of meat") or just 三磅 (sān bàng, "three pounds", never *三个磅 sān ge bàng).[24] Units of currency behave similarly: for example, 十元 (shí yuán, "ten yuan"), which is short for (for example) 十元人民币 (shí yuán rénmínbì, "ten units of renminbi"). Other proposed types of mass-classifiers include "collective"[25][note 10] mass-classifiers, such as 一群人 (yì qún rén, "a crowd of people"), which group things less precisely; and "container"[26] mass-classifiers which group things by containers they come in, as in 一碗粥 (yì wǎn zhōu, "a bowl of porridge") or 一包糖 (yì bāo táng, "a bag of sugar").

The difference between count-classifiers and mass-classifiers can be described as one of quantifying versus categorizing: in other words, mass-classifiers create a unit by which to measure something (i.e. boxes, groups, chunks, pieces, etc.), whereas count-classifiers simply name an existing item.[27] Most words can appear with both count-classifiers and mass-classifiers; for example, pizza can be described as both 一张比萨 (yì zhāng bǐsà, "one pizza", literally "one pie of pizza"), using a count-classifier, and as 一块比萨 (yí kuài bǐsà, "one piece of pizza"), using a mass-classifier. In addition to these semantic differences, there are differences in the grammatical behaviors of count-classifiers and mass-classifiers;[28] for example, mass-classifiers may be modified by a small set of adjectives (as in 一大群人 yí dà qún rén, "a big crowd of people"), whereas count-classifiers usually may not (for example, *一大个人 yí dà ge rén is never said for "a big person"; instead the adjective must modify the noun: 一个大人 yí ge dà rén).[29] Another difference is that count-classifiers may often be replaced by a "general" classifier 个 (個), gè with no apparent change in meaning, whereas mass-classifiers may not.[30] Syntacticians Lisa Cheng and Rint Sybesma propose that count-classifiers and mass-classifiers have different underlying syntactic structures, with count-classifiers forming "classifier phrases",[note 11] and mass-classifiers being a sort of relative clause that only looks like a classifier phrase.[31] The distinction between count-classifiers and mass-classifiers is often unclear, however, and other linguists have suggested that count-classifiers and mass-classifiers may not be fundamentally different. They posit that "count-classifier" and "mass-classifier" are the extremes of a continuum, with most classifiers falling somewhere in between.[32]

There is a set of "verbal classifiers" used specifically for counting the number of times an action occurs, rather than counting a number of items; this set includes 次 cì, 遍/徧 biàn, 回 huí, and 下 xià, which all roughly translate to "times".[33] For example, 我去过三次北京 (wǒ qù-guo sān cì Běijīng, I go-PAST three-CL Beijing, "I have been to Beijing three times").[34] These words can also form compound classifiers with certain nouns, as in 人次 rén cì "person-time", which can be used to count (for example) visitors to a museum in a year (where visits by the same person on different occasions are counted separately).

Another type of verbal classifier indicates the tool or implement used to perform the action. An example is found in the sentence 他踢了我一脚 tā tī le wǒ yī jiǎo "he kicked me", or more literally "he kicked me one foot". The word 脚 jiǎo, which usually serves as a simple noun meaning "foot", here functions as a verbal classifier reflecting the tool (namely the foot) used to perform the kicking action.

While mass-classifiers do not necessarily bear any semantic relationship to the noun with which they are used (e.g. box and book are not related in meaning, but one can still say "a box of books"), count-classifiers do.[31] The precise nature of that relationship, however, is not certain, since there is so much variability in how objects may be organized and categorized by classifiers. Accounts of the semantic relationship may be grouped loosely into categorical theories, which propose that count-classifiers are matched to objects solely on the basis of inherent features of those objects (such as length or size), and prototypical theories, which propose that people learn to match a count-classifier to a specific prototypical object and to other objects that are like that prototype.[39]

The categorical, "classical"[40] view of classifiers was that each classifier represents a category with a set of conditions; for example, the classifier 条 (條) tiáo would represent a category defined as all objects that meet the conditions of being long, thin, and one-dimensional—and nouns using that classifier must fit all the conditions with which the category is associated. Some common semantic categories into which count-classifiers have been claimed to organize nouns include the categories of shape (long, flat, or round), size (large or small), consistency (soft or hard), animacy (human, animal, or object),[41] and function (tools, vehicles, machines, etc.).[42]

A mule
骡子, luózi

A donkey
驴子, lǘzi

James Tai and Wang Lianqing found that the horse classifier 匹 pǐ is sometimes used for mules and camels, but rarely for the less "horse-like" donkeys, suggesting that the choice of classifiers is influenced by prototypal closeness.[43]

On the other hand, proponents of prototype theory propose that count-classifiers may not have innate definitions, but are associated with a noun that is prototypical of that category, and nouns that have a "family resemblance" with the prototype noun will want to use the same classifier.[note 13] For example, horse in Chinese uses the classifier 匹 pǐ, as in 三匹马 (sān pǐ mǎ, "three horses")—in modern Chinese the word 匹 has no meaning. Nevertheless, nouns denoting animals that look like horses will often also use this same classifier, and native speakers have been found to be more likely to use the classifier 匹 the closer an animal looks to a horse.[43] Furthermore, words that do not meet the "criteria" of a semantic category may still use that category because of their association with a prototype. For example, the classifier 颗 (顆) kē is used for small round items, as in 一颗子弹 (yì kē zǐdàn, "one bullet"); when words like 原子弹 (yuánzǐdàn, "atomic bomb") were later introduced into the language they also used this classifier, even though they are not small and round—therefore, their classifier must have been assigned because of the words' association with the word for bullet, which acted as a "prototype".[44] This is an example of "generalization" from prototypes: Erbaugh has proposed that when children learn count-classifiers, they go through stages, first learning a classifier-noun pair only (such as 条鱼 tiáo yú, CL-fish), then using that classifier with multiple nouns that are similar to the prototype (such as other types of fish), then finally using that set of nouns to generalize a semantic feature associated with the classifier (such as length and flexibility) so that the classifier can then be used with new words that the person encounters.[45]

Some classifier-noun pairings are arbitrary, or at least appear to modern speakers to have no semantic motivation.[46] For instance, the classifier 部 bù may be used for movies and novels, but also for cars[47] and telephones.[48] Some of this arbitrariness may be due to what linguist James Tai refers to as "fossilization", whereby a count-classifier loses its meaning through historical changes but remains paired with some nouns. For example, the classifier 匹 pǐ used for horses is meaningless today, but in Classical Chinese may have referred to a "team of two horses",[49] a pair of horse skeletons,[50] or the pairing between man and horse.[51][note 14] Arbitrariness may also arise when a classifier is borrowed, along with its noun, from a dialect in which it has a clear meaning to one in which it does not.[52] In both these cases, the use of the classifier is remembered more by association with certain "prototypical" nouns (such as horse) rather than by understanding of semantic categories, and thus arbitrariness has been used as an argument in favor of the prototype theory of classifiers.[52] Gao and Malt propose that both the category and prototype theories are correct: in their conception, some classifiers constitute "well-defined categories", others make "prototype categories", and still others are relatively arbitrary.[53]

In addition to the numerous "specific" count-classifiers described above,[note 15] Chinese has a "general" classifier 个 (個), pronounced gè in Mandarin.[note 16] This classifier is used for people, some abstract concepts, and other words that do not have special classifiers (such as 汉堡包 hànbǎobāo "hamburger"),[54] and may also be used as a replacement for a specific classifier such as 张 (張) zhāng or 条 (條) tiáo, especially in informal speech. In Mandarin Chinese, it has been noted as early as the 1940s that the use of 个 is increasing and that there is a general tendency towards replacing specific classifiers with it.[55] Numerous studies have reported that both adults and children tend to use 个 when they do not know the appropriate count-classifier, and even when they do but are speaking quickly or informally.[56] The replacement of a specific classifier with the general 个 is known as classifier neutralization[57] ("量词个化" in Chinese, literally "classifier 个-ization"[58]). This occurs especially often among children[59] and aphasics (individuals with damage to language-relevant areas of the brain),[60][61] although normal speakers also neutralize frequently. It has been reported that most speakers know the appropriate classifiers for the words they are using and believe, when asked, that those classifiers are obligatory, but nevertheless use 个 without even realizing it in actual speech.[62] As a result, in everyday spoken Mandarin the general classifier is "hundreds of times more frequent"[63] than the specialized ones.

Nevertheless, 个 has not completely replaced other count-classifiers, and there are still many situations in which it would be inappropriate to substitute it for the required specific classifier.[55] There may be specific patterns behind which classifier-noun pairs may be "neutralized" to use the general classifier, and which may not. Specifically, words that are most prototypical for their categories, such as paper for the category of nouns taking the "flat/square" classifier 张 (張) zhāng, may be less likely to be said with a general classifier.[64]

A painting may be referred to with the classifiers 张 (張) zhāng and 幅 fú; both phrases have the same meaning, but convey different stylistic effects.[65]

Depending on the classifier used, the noun 楼 lóu could be used to refer to either this building, as in 一座楼 (yí zuò lóu "one building"), or the floors of the building, as in 二十层楼 (èrshí céng lóu, "twenty floors").[66]

It is not the case that every noun is only associated with one classifier. Across dialects and speakers there is great variability in the way classifiers are used for the same words, and speakers often do not agree which classifier is best.[67] For example, for cars some people use 部 bù, others use 台 tái, and still others use 辆 (輛) liàng; Cantonese uses 架 gaa3. Even within a single dialect or a single speaker, the same noun may take different measure words depending on the style in which the person is speaking, or on different nuances the person wants to convey (for instance, measure words can reflect the speaker's judgment of or opinion about the object[68]). An example of this is the word for person, 人 rén, which uses the measure word 个 (個) gè normally, but uses the measure 口 kǒu when counting number of people in a household, 位 wèi when being particularly polite or honorific, and 名 míng in formal, written contexts;[69] likewise, a group of people may be referred to by massifiers as 一群人 (yì qún rén, "a group of people") or 一帮人 (yì bāng rén, "a gang/crowd of people"): the first is neutral, whereas the second implies that the people are unruly or otherwise being judged poorly.[70]

Some count-classifiers may also be used with nouns that they are not normally related to, for metaphorical effect, as in 一堆烦恼 (yì duī fánnǎo, "a pile of worries/troubles").[71] Finally, a single word may have multiple count-classifiers that convey different meanings altogether—in fact, the choice of a classifier can even influence the meaning of a noun. By way of illustration, 三节课 sān jié kè means "three class periods" (as in "I have three classes today"), whereas 三门课 sān mén kè means "three courses" (as in "I signed up for three courses this semester"), even though the noun in each sentence is the same.[66]

An oracle bone inscription from the Shāng dynasty. Such inscriptions provide some of the earliest examples of the number phrases that may have eventually spawned Chinese classifiers.

Historical linguists have found that phrases consisting of nouns and numbers went through several structural changes in Old Chinese and Middle Chinese before classifiers appeared in them. The earliest forms may have been Number – Noun, like English (i.e. "five horses"), and the less common Noun – Number ("horses five"), both of which are attested in the oracle bone scripts of Pre-Archaic Chinese (circa 1400 BCE to 1000 BCE).[79] The first constructions resembling classifier constructions were Noun – Number – Noun constructions, which were also extant in Pre-Archaic Chinese but less common than Number – Noun. In these constructions, sometimes the first and second nouns were identical (N1 – Number – N1, as in "horses five horses") and other times the second noun was different, but semantically related (N1 – Number – N2). According to some historical linguists, the N2 in these constructions can be considered an early form of count-classifier and has even been called an "echo classifier"; this speculation is not universally agreed on, though.[80] Although true count-classifiers had not appeared yet, mass-classifiers were common in this time, with constructions such as "wine – six – yǒu" (the word 酉 yǒu represented a wine container) meaning "six yǒu of wine".[80] Examples such as this suggest that mass-classifiers predate count-classifiers by several centuries, although they did not appear in the same word order as they do today.[81]

It is from this type of structure that count-classifiers may have arisen, originally replacing the second noun (in structures where there was a noun rather than a mass-classifier) to yield Noun – Number – Classifier. That is to say, constructions like "horses five horses" may have been replaced by ones like "horses five CL", possibly for stylistic reasons such as avoiding repetition.[82] Another reason for the appearance of count-classifiers may have been to avoid confusion or ambiguity that could have arisen from counting items using only mass-classifiers—i.e. to clarify when one is referring to a single item and when one is referring to a measure of items.[83]

Historians agree that at some point in history the order of words in this construction shifted, putting the noun at the end rather than beginning, like in the present-day construction Number – Classifier – Noun.[84] According to historical linguist Alain Peyraube, the earliest occurrences of this construction (albeit with mass-classifiers, rather than count-classifiers) appear in the late portion of Old Chinese (500 BCE to 200 BCE). At this time, the Number – Mass-classifier portion of the Noun – Number – Mass-classifier construction was sometimes shifted in front of the noun. Peyraube speculates that this may have occurred because it was gradually reanalyzed as a modifier (like an adjective) for the head noun, as opposed to a simple repetition as it originally was. Since Chinese generally places modifiers before modified, as does English, the shift may have been prompted by this reanalysis. By the early part of the Common Era, the nouns appearing in "classifier position" were beginning to lose their meaning and become true classifiers. Estimates of when classifiers underwent the most development vary: Wang Li claims their period of major development was during the Han dynasty (206 BCE – 220 CE),[85] whereas Liu Shiru estimates that it was the Northern and Southern dynasties period (420 – 589 CE),[86] and Peyraube chooses the Tang dynasty (618 – 907 CE).[87] Regardless of when they developed, Wang Lianqing claims that they did not become grammatically mandatory until sometime around the 11th century.[88]

Classifier systems in many nearby languages and language groups (such as Vietnamese and the Tai languages) are very similar to the Chinese classifier system in both grammatical structure and the parameters along which some objects are grouped together. Thus, there has been some debate over which language family first developed classifiers and which ones then borrowed them—or whether classifier systems were native to all these languages and developed more through repeated language contact throughout history.[89]

Most modern count-classifiers are derived from words that originally were free-standing nouns in older varieties of Chinese, and have since been grammaticalized to become bound morphemes.[90] In other words, count-classifiers tend to come from words that once had specific meaning but lost it (a process known as semantic bleaching).[91] Many, however, still have related forms that work as nouns all by themselves, such as the classifier 带 (帶) dài for long, ribbon-like objects: the modern word 带子 dàizi means "ribbon".[71] In fact, the majority of classifiers can also be used as other parts of speech, such as nouns.[92] Mass-classifiers, on the other hand, are more transparent in meaning than count-classifiers; while the latter have some historical meaning, the former are still full-fledged nouns. For example, 杯 (bēi, cup), is both a classifier as in 一杯茶 (yì bēi chá, "a cup of tea") and the word for a cup as in 酒杯 (jiǔbēi, "wine glass").[93]

It was not always the case that every noun required a count-classifier. In many historical varieties of Chinese, use of classifiers was not mandatory, and classifiers are rare in writings that have survived.[94] Some nouns acquired classifiers earlier than others; some of the first documented uses of classifiers were for inventorying items, both in mercantile business and in storytelling.[95] Thus, the first nouns to have count-classifiers paired with them may have been nouns that represent "culturally valued" items such as horses, scrolls, and intellectuals.[96] The special status of such items is still apparent today: many of the classifiers that can only be paired with one or two nouns, such as 匹 pǐ for horses[note 18] and 首 shǒu for songs or poems, are the classifiers for these same "valued" items. Such classifiers make up as much as one-third of the commonly used classifiers today.[19]

Classifiers did not gain official recognition as a lexical category (part of speech) until the 20th century. The earliest modern text to discuss classifiers and their use was Ma Jianzhong's 1898 Ma's Basic Principles for Writing Clearly (马氏文通).[97] From then until the 1940s, linguists such as Ma, Wang Li, and Li Jinxi treated classifiers as just a type of noun that "expresses a quantity".[85] Lü Shuxiang was the first to treat them as a separate category, calling them "unit words" (单位词 dānwèicí) in his 1940s Outline of Chinese Grammar (中国文法要略) and finally "measure words" (量词 liàngcí) in Grammar Studies (语法学习). He made this separation based on the fact that classifiers were semantically bleached, and that they can be used directly with a number, whereas true nouns need to have a measure word added before they can be used with a number.[98] After this time, other names were also proposed for classifiers: Gao Mingkai called them "noun helper words" (助名词 zhùmíngcí), Lu Wangdao "counting markers" (计标 jìbiāo), and Japanese linguist Miyawaki Kennosuke called them "accompanying words" (陪伴词 péibàncí).[99] In the Draft Plan for a System of Teaching Chinese Grammar  adopted by the People's Republic of China in 1954, Lü's "measure words" (量词 liàngcí) was adopted as the official name for classifiers in China.[100] This remains the most common term in use today.[12]

Historically, 个 gè was not always the general classifier. Some believe it was originally a noun referring to bamboo stalks, and gradually expanded in use to become a classifier for many things with "vertical, individual, [or] upright qualit[ies]",[101] eventually becoming a general classifier because it was used so frequently with common nouns.[102] The classifier gè is actually associated with three different homophonous characters: 个, 個 (used today as the traditional-character equivalent of 个), and 箇. Historical linguist Lianqing Wang has argued that these characters actually originated from different words, and that only 箇 had the original meaning of "bamboo stalk".[103] 个, he claims, was used as a general classifier early on, and may have been derived from the orthographically similar 介 jiè, one of the earliest general classifiers.[104] 箇 later merged with 介 because they were similar in pronunciation and meaning (both used as general classifiers).[103] Likewise, he claims that 個 was also a separate word (with a meaning having to do with "partiality" or "being a single part"), and merged with 个 for the same reasons as 箇 did; he also argues that 個 was "created", as early as the Han dynasty, to supersede 个.[105]

Nor was 个 the only general classifier in the history of Chinese. The aforementioned 介 jiè was being used as a general classifier before the Qin dynasty (221 BCE); it was originally a noun referring to individual items out of a string of connected shells or clothes, and eventually came to be used as a classifier for "individual" objects (as opposed to pairs or groups of objects) before becoming a general classifier.[106] Another general classifier was 枚 méi, which originally referred to small twigs. Since twigs were used for counting items, 枚 became a counter word: any items, including people, could be counted as "one 枚, two 枚", etc. 枚 was the most common classifier in use during the Northern and Southern dynasties period (420–589 CE),[107] but today is no longer a general classifier, and is only used rarely, as a specialized classifier for items such as pins and badges.[108] Kathleen Ahrens has claimed that 隻 (zhī in Mandarin and jia in Taiwanese), the classifier for animals in Mandarin, is another general classifier in Taiwanese and may be becoming one in the Mandarin spoken in Taiwan.[109]

Northern dialects tend to have fewer classifiers than southern ones. 個 ge is the only classifier found in the Dungan language. All nouns could have just one classifier in some dialects, such as Shanghainese (Wu), the Mandarin dialect of Shanxi, and Shandong dialects. Some dialects such as Northern Min, certain Xiang dialects, Hakka dialects, and some Yue dialects use 隻 for the noun referring to people, rather than 個.[110]