A word between the number and the thing
In English you can say “three cats” or “three books” and nothing sits between the number and the noun. In Chinese, Japanese and Korean the number usually cannot attach directly to the noun. A counter word, called a classifier or measure word, goes in between. These words are also called numeral classifiers, and they are a standard part of counting in all three languages.
The English phrase “three head of cattle” or “two sheets of paper” gives a rough idea, but in English such words are used only in some contexts. In these three languages the classifier is a routine requirement.
Chinese
In Mandarin Chinese the pattern is number, classifier, noun. The general-purpose classifier is “ge”, as in “yi ge ren” (“one person”). Many nouns take more specific classifiers that reflect something about the thing counted:
- “tiao” is used for long, flexible things such as fish, rivers and roads
- “zhang” is used for flat things such as paper, tables and tickets
- “ben” is used for bound volumes such as books
- “zhi” is used for many animals and for one of a pair of items
So “one fish” is “yi tiao yu” and “one piece of paper” is “yi zhang zhi”. The classifier is not decoration. It must be there for the sentence to be correct, although “ge” is often used loosely in casual speech where a more specific word would be expected. The numeral for “two” also changes form in this position: Mandarin uses “liang” rather than “er” before a classifier.
Japanese
Japanese uses classifiers too, and the number often comes after the noun, as in “neko ga ni-hiki” (“two cats”), though a number-classifier phrase can also come before the noun. Examples include:
- “hon” for long, thin objects such as pencils, bottles and trees
- “mai” for flat objects such as sheets, shirts and plates
- “hiki” for small animals
- “satsu” for books and other bound volumes
- “nin” for people
- “dai” for machines and vehicles
Japanese classifiers also bring sound changes. The “hon” counter becomes “-pon” or “-bon” after some numbers: “ippon” (one), “nihon” (two), “sanbon” (three). People have irregular forms: one person is “hitori” and two people are “futari”, and only from three do the regular “-nin” forms appear. These sound changes interact with the two sets of readings for numbers that Japanese uses, a topic that belongs with the Japanese number readings.
A folk explanation often given for why rabbits are counted with “wa”, the counter used for birds, is that eating four-legged animals was once frowned on in a Buddhist context, so rabbits were treated as birds in speech. This is a popular story and its accuracy is uncertain, so it should be treated as a tradition about the word, not a settled fact.
Korean
Korean is similar in structure. A numeral, a classifier and a noun are combined, and the classifier depends on the type of thing counted. Common ones include “myeong” for people (often with a polite “bun” variant), “mari” for animals, “gae” as a general counter, “gwon” for books and “jang” for flat objects such as paper and tickets.
Korean has an additional layer, because it has two sets of numbers: a native series and a Sino-Korean series borrowed from Chinese. Most common classifiers go with the native numbers, which often change form before the classifier. “One person” is “han myeong” and “two animals” is “du mari”. Other units, such as dates, money and some measures, use the Sino-Korean series. Which set goes with which classifier mostly has to be learned.
How classifiers differ from English measure words
English has measure words, such as “a cup of tea”, “a sheet of paper” and “a pair of shoes”. These are needed mostly with mass nouns, things that cannot be counted directly. Countable nouns like “cat” combine with a number directly in English, while in the East Asian languages most nouns need a classifier.
Classifier languages are common outside East Asia as well. Thai, Vietnamese, Bengali and many Mayan languages are examples, and each organises its classifiers differently. The classifiers in one language may sort objects by shape, animacy, size or function, and the categories do not always match those of another. This variety is one reason that classifier systems appeal to linguists.
Where classifiers may have come from
Many classifiers appear to be old nouns that became grammatical markers. Linguists generally think that classifiers developed from nouns used in counting phrases, though the details differ by language and many individual etymologies are uncertain.
It is easy to read too much into these categories. A classifier tells you how a language groups objects for counting, not how speakers view the world. A speaker who uses “tiao” for fish and roads is not necessarily thinking of them as the same kind of thing; it is a grammatical habit and its history is a matter for linguists.
Numbers and counters in everyday life
Classifiers show up wherever things are counted in these languages. Shops, menus and signs depend on them, and learners often find the choice of classifier harder than the numbers themselves. Beyond language, numbers have their own cultural associations, including the sound-based beliefs about Chinese number homophones, where the reading of a number matters because of similar-sounding words. The sound of a number and the counter after it work independently, but both reflect how deeply counting is embedded in each language.
There is a loose parallel with East Asian calculation, which relied on physical counters on the abacus, in a history that runs alongside the verbal counters discussed here. The two are separate traditions, and no direct link between them is claimed.