Laryngeal features

Three features characterize the state of the glottis:

spread glottis (s.g.): the vocal folds are spread far apart.

constricted glottis (c.g.): the vocal folds are tightly constricted.

voice (voi): the vocal folds vibrate.

Voiced sounds are [+voice]. The feature [spread glottis] describes aspirated obstruents ([ph ], [bh ]) and breathy sonorants ([m̤ ], [a̤]); [constricted glottis] describes implosives ([ɓ]), ejective obstruents ([p’]), and laryngealized sonorants ([m̰], [a̰]).

How to distinguish implosives from ejectives is not entirely obvious, but the standard answer is that ejectives are [-voice] and implosives are [+voice]. There are two problems with this. One is that implosives do not generally pattern with other [+voiced] consonants in phonological systems, especially in how consonants affect tone (voiced consonants, but typically not implosives, may lower following tones). The second is that Ngiti and Lendu have both voiced and voiceless implosives. The languages lack ejectives, which raises the possibility that voiceless implosives are phonologically [-voice, +c.g.], which is exactly the specification given to ejective consonants. You may wonder how [-voice, +c.g.] can be realized as an ejective in languages like Navajo, Tigre or Lushootseed, and as a voiceless implosive in Ngiti or Lendu. This is possible because feature values give approximate phonetic descriptions, not exact ones. The Korean “fortis” consonants, found in [k’ata] ‘peel (noun),’ [ak’i] ‘musical instrument,’ or [alt’a] ‘be ill,’ are often described as glottalized, and phonetic studies have shown that they are produced with glottal constrictions: thus they would be described as [-voice, +c.g.]. Nevertheless, they are not ejectives. Similarly, Khoekhoe (Nama) has a contrast between plain clicks ([!àḿ] ‘deep’) and glottalized ones ([!’ám̋] ‘kill’), but the glottalized clicks realize the feature [+c.g.] as a simple constriction of the glottis, not involving an ejective release.

The usual explanation for the difference between ejectives in Navajo and glottalized nonejective consonants in Korean or Khoekhoe is that they have the same phonological specifications, [-voice, +c.g.], but realize the features differently due to language-specific differences in principles of phonetic implementation. This is an area of feature theory where more research is required.

The representations of laryngeal contrasts in consonants are given below.