A nasal vowel is a vowel that is produced with a lowering of the soft palate (or velum) so that the air flow escapes through the nose and the mouth simultaneously, as in the French vowel or Amoy . By contrast, oral vowels are produced without nasalization. In a stricter sense, nasal vowels shall not be confused with nasalised vowels.
Nasalised vowels are vowels under the influence of neighbouring sounds. For instance, the [æ] of the word hand is affected by the following nasal consonant. In most languages, vowels adjacent to nasal consonants are produced partially or fully with a lowered velum in a natural process of assimilation and are therefore technically nasal, but few speakers would notice. That is the case in English: vowels preceding nasal consonants are nasalized, but there is no phonemic distinction between nasal and oral vowels (and all vowels are considered phonemically oral).
However, the words "huh?" and "uh-huh" are pronounced with a nasal vowel, as is the negative "unh-unh".
The nasality of nasal vowels, however, is a distinctive feature of certain languages. In other words, a language may contrast oral vowels and nasalised vowels phonemically. Linguists make use of minimal pairs to decide whether or not the nasality is of linguistic importance. In French, for instance, nasal vowels are distinct from oral vowels, and words can differ by this vowel quality. The words beau /bo/ "beautiful" and bon /b/ "good" are a minimal pair that contrasts primarily the vowel nasalization, even if the // from bon is slightly more open.
Portuguese behaves similarly with pairs as vim /v?j?/ "I came" and vi /vi/ "I saw", except /?j?/ and /i/ are of same vowel height. Portuguese also allows nasal diphthongs that contrast with their oral counterparts, like the pair mau /'maw/ "bad" and mão /'mw?/ "hand".
Although there are French loanwords into English with nasal vowels like croissant ['k?w?:s], there is no expectation that an English speaker would nasalize the vowels to the same extent that French or Portuguese speakers do. Likewise, pronunciation keys in English dictionaries do not always indicate nasalization of French loanwords.
A few languages, such as Palantla Chinantec, contrast lightly nasalized and heavily nasalized vowels. They may be contrasted in print by doubling the IPA diacritic for nasalization: ⟨?⟩ vs ⟨⟩. Bickford & Floyd (2006) combine the tilde with the ogonek: ⟨?⟩ vs ⟨⟩. (The ogonek is sometimes used in an otherwise IPA transcription to avoid conflict with tone diacritics above the vowels.)
Rodney Sampson described a three-stage historical account, explaining the origin of nasal vowels in modern French. The notation of Terry and Webb will be used below, where V, N, and ? (with a tilde above) represent oral vowel, nasal consonant, and nasal vowel, respectively.
|Stage 1||Stage 2||Stage 3|
|ca. 13th||ca. 14th-16th||ca. 17th-18th|
|vend [v?nt], [vnt]||[v(n)t]||[v]|
In the Old French period, vowels become nasalised under the regressive assimilation, as VN > ?N. In the Middle French period, the realisation of the nasal consonant became variable, as VN > ?(N). As the language evolves into its modern form, the consonant is no longer realised, as ?N > ?.
In other cases, they are indicated by diacritics. In the International Phonetic Alphabet, nasal vowels are denoted by a tilde over the symbol for the vowel. The same practice can be found in Portuguese marking with a tilde in diphthongs (e.g. põe) and for words ending in // (e.g. manhã, ímã). While the tilde is also used for this purpose in Paraguayan Guaraní, phonemic nasality is indicated by a diaeresis ( ¨ ) in the standardized orthographies of most varieties of Tupí-Guaraní spoken in Bolivia. Polish, Navajo, and Elfdalian use a hook under the letter, called an ogonek, as in ?, ?. The Pe?h-?e-j? romanization of Taiwanese Hokkien and Amoy uses a superscript n (a?, e?, ...).
The Brahmic scripts used for most Indic languages mark nasalization with the anusv?ra (? , homophonically used for homorganic nasalization in a consonant cluster following the vowel) or the anun?sika (?) diacritic (and its regional variants).
Nasalization in Nastaliq-based Arabic scripts of Urdu (as well as Western Punjabi) is indicated by placing after the vowel a dotless form of the Arabic letter n?n (?) or the letter marked with the maghn?na diacritic: respectively ? (always occurring word finally) or , called "n?n ghunna". Nasalized vowels occur in Classical Arabic but not in contemporary speech or Modern Standard Arabic. There is no orthographic way to denote the nasalization, but it is systematically taught as part of the essential rules of tajwid, used to read the Qur'an. Nasalization occurs in recitation, usually when a final n?n is followed by a y (?).
These languages use phonemic nasal vowels: