类型学与历时语言学数据库
看世界各语言如何共享、差异、变化。
Language Patterns记录世界各语言反复出现的策略、路径与循环——从互不相关的语言如何表达「有」，到否定词如何在数百年间不断重建。
How do languages say “there is”?
“The water is in the river” takes the water as given and tells you where it is. “There is water in the river” reverses that: the river is the setting and the water is the news. Saying that something exists therefore reuses the machinery for saying where it is, and languages differ in how much they change along the way — some leave the locational predicate untouched, some add a placeholder subject, some borrow the possessive verb, and some keep a predicate for existence alone. The choice often splits by animacy and by number.
How do languages say “I have X”?
Some languages use a transitive HAVE verb; others say “at me, X is”, “to me, X is”, or “I-with X”. The same language often switches strategy with what is possessed — inanimate, animate, kin, body-parts.
Do “my hand” and “my house” use the same grammar?
Hands and mothers come with a possessor built in; houses and books change hands. Grammars pick this up: some force a possessive onto body-part and kin nouns or give them special possessed forms, some route acquired things through a classifier, Mandarin lets its linker drop exactly with kin, and much of Europe moves the body-part possessor onto the verb as a dative.
Where does “the man’s head” carry its mark?
Two nouns get tied together in five ways: a genitive on the possessor (Latin, Russian, English ’s), a person affix on the possessed noun (Hungarian, Ainu, Navajo), both at once (Turkish, Quechua), a free particle between the two (Japanese の, Mandarin 的, Romance de), or nothing beyond word order (Welsh, Indonesian).
How do languages say “I don’t have X”?
To express absence of possession, languages either negate a HAVE-verb, add a negative determiner to the object, or reframe the situation as one of non-existence or absence — with the possessor as topic, location, or dative.
How do languages build the paradigm of someone, anyone, no one, and everyone?
Most often these words share one interrogative base (who?) plus a small particle — “also”, “every”, “some”, “any” — that sets the reading. A few languages, English among them, lexicalise the series instead.
In what order do languages arrange subject, object, and verb?
Two of the six logically possible orders dominate the world’s languages — SOV (Japanese, Hindi, Turkish) and SVO (English, Mandarin, Romance). The rest are rare, or set by information structure rather than a fixed template.
Do adpositions come before or after the noun phrase?
Adposition order tracks verb–object order as a strong tendency — after the noun in OV languages (Japanese tēburu no ue ni), before it in VO (English on the table). Exceptions recur across families: Persian and Tajik combine OV order with prepositions, while SVO-leaning Finnish keeps its Uralic postpositions.
Which verb makes a machine start working?
A lamp, an air conditioner, a tap, recorded music, an alarm clock, a vacuum cleaner and a washing machine each need a verb. Some languages reach for fire, some for opening or rotation, some for the plain act of putting a thing down. The choice also shifts from one thing to the next within a single language: Japanese attaches a light and hangs music, where Korean kindles the light and rotates the washing machine.
How does a language say an action was done for someone?
English hangs a beneficiary on the end of the clause with “for”, or slips it in front of the object: cook a meal for her, cook her a meal. Elsewhere the verb itself carries the news. Swahili lengthens the stem, Tagalog rebuilds the clause around the person who profits, and across East and Southeast Asia into West Africa a second verb meaning “give” follows the first. Japanese splits that second verb three ways, so one act of cooking is worded differently depending on who ends up better off — and uses the passive when the answer is nobody.
How does negation rebuild itself over time?
A preverbal negator weakens and is reinforced by an extra element, often a minimizer like “step” or “thing”. The reinforcer is reanalysed as the negator itself, then weakens in turn, and the cycle starts again.
How does a word for “there is no” turn into plain “not”?
A dedicated negative existential (“there is no X”) arises beside the ordinary negator, then spreads from existential sentences into verbal negation until it becomes the standard negator itself. Croft distinguishes six synchronic types — A, A~B, B, B~C, C, C~A — and the languages below sit at different points along the drift.
How does “listen” come to mean “obey”?
Verbs of hearing and listening repeatedly grow a sense of obeying. The pattern recurs from Mandarin tīnghuà and Japanese iu koto o kiku to Latin oboedīre, built on audīre “to hear”, behind French obéir and English obey.
How does “see” come to mean “try”?
Verbs of seeing repeatedly become markers of attempting. Japanese 食べてみる “eat and see”, Korean 먹어 보다, Mandarin 試試看, and the Turkic converb + kör- constructions all say “do it and see” for “try doing it”; Voinov (2013) traces the co-lexification across Eurasia, New Guinea, India, and West Africa.