In linguistics, heterography is a property of a written language, such that it lacks a 1-to-1 correspondence between the written symbols and the sounds of the spoken language.[1] Its opposite is homography, which is the property of a language such that written symbols of its written form and the sounds of its spoken form have a 1-to-1 correspondence.[2]
The orthography of the English language is, according to Larry Trask, a "spectacular example" of heterography. But most European languages exhibit it to some extent. Finnish is "very close" to being a systematically homographic language.[1][2] A phonemic transcription (such as a transcription of phonemes that uses the International Phonetic Alphabet, for example) is, by its nature, homographic, also.[2]
The degree of heterography of a language is a factor in how difficult it is for person to learn to read that language, with highly heterographic orthographies being more difficult to learn than more homographic ones. Many people have espoused the point of view that the extreme heterographic nature of English is a disadvantage in several respects. These include, for example, Dr. Kiyoshi Makita writing in the July 1968 issue of the American Journal of Orthopsychiatry, who attributes the rarity of dyslexia amongst Japanese children to the fact that Japanese is highly homographic language.[3]
| Written forms | |||
|---|---|---|---|
| same | different | ||
| Sounds | same | Homophonic homographs | Homophonic heterographs |
| different | Heterophonic homographs | Heterophonic heterographs | |
Certainly, confusion between heterographic homophonic words (so-called homophonic heterographs, a.k.a. heterographic homophones), such as "piece" and "peace" is one of the symptoms of surface dyslexia, one of the forms of dyslexia.[4]
Other homophonic heterographs in English include "right", "rite", and "wright", and "there", "their" and "they're".[5] In French, examples include "sain" and "saint".[6]
Heterophonic homographs (also known as homographic heterophones) are, in contrast, words whose spoken sounds differ but whose written forms are the same. English has a few hundred heterophonic homographs,[7] examples of these latter include "read" ("will read" vs "has read").[5][7]
The two aforementioned classes of words, along with a third class (homophonic homographs — words with different meanings whose written and spoken forms are both the same, such as "bank" in English and "杜鹃" in Chinese) are the three classes of lexical ambiguities in all languages. (They are marked in green in the key on the right.)[8]
Chinese has many words that are both homophonic and homotonic. Distinctions are made between such words using heterography.[9] Homophonic heterographs are very frequent in Chinese, whereas heterophonic homographs are not.[8] In contrast, homographic heterophony is one of the most salient characteristics of English orthography, with the "-ough" in "though", "tough", "through", and "thought" being homographic but greatly heterophonic.[10]
French is more heterographic than English. Whilst it, too, has homographic heterophony, such as the "-ars" in "Mars", "jars", and "gars", its most prominent irregularity is heterographic, namely heterographic homophony. Witness, for example, the identical pronunciation (in some regional dialects) of "-eng", "-empt", "-amp", and "-ans" in "hareng", "champ", "exempt", and "dans".[10]