In studying language, it is often pointed out that the connection between any word and what it signifies is arbitrary; that is, there is no a priori, compelling reason why, for instance, the domesticated four-legged canine animal should be called dog—it could very well have been pig. It was coincidental that this animal was named dog and it was through centuries of conventional use that dog became the linguistic sign in English to refer to this actual animal. The principle behind the arbitrariness of meaning-symbol connection holds up across languages. Nevertheless, there is a class of words in languages where the sound-meaning relationship shows certain connections. Such is the case with onomatopoeia, which is defined in the American Heritage Dictionary as words that "imitate the sounds associated with the objects or actions they refer to," such as "buzz" or "murmur."
Onomatopoeia, which might be defined as mimetic language, thus refers to the phenomenon where the pronunciation of a word suggests its actual meaning. In English we may describe the sound of a rock falling into water as splash—this is onomatopoeic. Examples abound from other languages in the same vein. This phenomenon of onomatopoeia illustrates an important aspect of language, for it is language that imposes a constraint on the description, that is different language begets different onomatopoeia for the 'same' sound. So the same sound is described differently in different languages. Take a recording of a dog barking. Japanese might describe this as wanwan, English ruffruff, Chinese wangwang, Russian guffguff, and so forth. Why the variety? Do humans not hear the same sound in the same way? The sound we hear is arguably identical but speakers of different languages describe it differently, each influenced by the sound structure of the language in question.
In English onomatopoeic words may be limited to certain language context, such as comic books and graphic, animated telling of an event. In Japanese, however, onomatopoeia constitutes a large class of words, numbering in the thousands. According to one conservative estimate, Japanese makes use of onomatopoeia three times as often as English. In Japanese, onomatopoeia is used in all kinds of prose and speech, formal or informal, whenever a precise, apt description is demanded. In Japanese grammar onomatopoeic words also function adverbially, so that it can be followed by X-(t)to 'saying X, making the sound X' and make an adverbial phrase modifying the predicate directly to signify the final state of action as in kushakusha tto marumeta 'crumpled (paper)' or followed by an adverb making particle ni as in kushakusha ni 'into a crumpled wad.'
We also notice that almost all onomatopoeic words are reduplicated because they are descriptive of repeated, rhythmical sounds (but see, shikkari 'firmly' hatto 'all of a sudden'). Across languages reduplication has semantic commonalty, used to show plurality, repetition, and higher intensity of a phenomenon. In the examples below, we can see this principle is at work in Japanese.
There are two phenomena surrounding the use of onomatopoeia. First is the question as to what the speakers of Japanese implicitly understand the tacit rules needed to interpret or generate new onomatopoeia. The wide applicability of these rules means that brand new onomatopoeic words made up by someone are be readily understood as conveying a certain sense. The second concerns what sort of activities Japanese onomatopoeia covers. For although true onomatopoeia usually refers to words imitating sounds occurring in nature, many onomatopoeic words in Japanese are capable of describing activities and states that do not involve sound.