Main Page | See live article | Alphabetical index


A vowel is a sound in spoken language that has a sounding voice (vocal sound) of its own, it is produced by comparatively open configuration of the vocal tract. Unlike a non-vowel (consonant), a vowel can be sounded on its own. A single vowel sound forms the basis of a syllable, although two adjacent vowel sounds can be blended together into a single syllable diphthong.

For those languages which use some form of the Roman alphabet, such as English, certain letters are identified as vowels because they are normally associated with vowel sounds.

In the English language, the vowel letters are A , E , I, O , U and sometimes Y. In English, the letter W by itself is not usually a vowel, but can form a diphthong with the vowels A, E, or O, and can serve as a vowel in a few Welsh-derived words like cwm and crwth. In old English V was used interchangeably with u and j was used interchangeably with i (particular in Roman numerals, e.g. vij). In certain other languages which use the Roman alphabet, consonants can serve as vowels: vrba is Czech for "willow" and slnko is Slovak for "sun".

There is necessarily not a direct one-to-one correspondence between the vowel sounds of a language and the vowel letters. Many languages that use a form of the Roman alphabet have more vowel sounds than can be represented by the standard vowel letters. In the case of English, the five primary vowel letters can represent both long and short vowel sounds (some of the long vowel sounds in English are actually diphthongs). Furthermore, in English some vowel sounds are represented by combinations of vowel letters, such as the ea in beat or the ou in such words as through or thought. Note that the consonants gh in these words are not part of the vowel sequence. While always silent in Standard English, they may be pronounced in other English dialects.

Other languages also attempt to overcome the limitation in the number of Roman vowel letters in similar ways. Many languages, like English, make extensive use of combinations of vowel letters to represent various sounds. However, it is also very common for languages to add diacritical marks to vowels, such as accents or umlauts, to represent the variety of possible vowel sounds. Some languages have also constructed additional vowels that are based on the standard Roman vowels, such as ś or Ý that are found in some of the Scandinavian languages.

Vowels support the neighbouring consonants, but often bear little information themselves (cn y rd ths?). This is especially true of some Afro-Asiatic languages, where vowels carry mostly inflections and create few minimal pairs that would get in the way of understanding. Because of this, some alphabets (called abjads) do not represent vowels at all.

The vowels of many languages, though, are more structurally important to the word. Sedang, for instance, a relative of Vietnamese, contrasts 55 different vowels; in some languages with few consonants, the vowels are also more important to the structure of the word. Pacific languages such as Maori and Fijian are of this sort.

The most common vowel system, in both natural languages and constructed languages, is this one:

u     i
o   e

The reason for this seems to be that it makes the most efficient use of phonological space. Latin and Ancient Greek had this vowel system; it is for this reason that the Cyrillic, Greek and Latin alphabets have five vowel letters.

Languages can make several distinctions (contrasts) on vowels:

All languages have at least two vowels; the Tshwizhyi and Abzhui dialects of Abkhaz contrasts only /a/ and /i/, with significant allophony.¹ A few languages, such as Navajo, have four-vowel systems that lack either i or u, but there is no known natural language without a.

Daniel Jones developed a system to describe vowels.

¹ Some linguists claim that it is possible to posit only one vowel in some Abkhaz dialects, though the general consensus seems to be that that is stretching things a bit.