LINGUIST List 11.8

Tue Jan 11 2000

Sum: Q:10.1720: Phonetics/Ejectives/Laryngealization

Editor for this issue: Naomi Ogasawara <>


  1. Johannes Reese, Summary for 10.1720: creaky/stiff voice/laryngealised vs. ejectives

Message 1: Summary for 10.1720: creaky/stiff voice/laryngealised vs. ejectives

Date: Fri, 04 Jan 1980 15:41:34 +0100
From: Johannes Reese <>
Subject: Summary for 10.1720: creaky/stiff voice/laryngealised vs. ejectives

Thanks to:

"Kimary Shahin" <>
Michael Jessen <>
"Mailbox" <>
"mark jones" <>
John Davis <>
Nick Reid <>
Natasha Warner <>
"Roy Dace" <>
Mike Maxwell <>
Ronald Cosper <Ronald.CosperSTMARYS.CA>

Some of the contributors pointed me to Ladefoged's book "The
sounds of the world's languages; unfortunately, it was this that
made me wonder about the resemblance of laryngealized sounds and
ejectives. They seem to be, though, phonetically totally
different. Some of those who answered argued that phonologically,
the difference is indeed irrelevant, as both features never
co-occur in the same language (Jessen's description: [+constricted
glottis, -voice] or [+checked, -voiced] in Jacobsonian terms).
Unlike ejectives, laryngealized consonants may be voiced.

Shahin, Jones, and Reid emphasize the articulatory nature of the
laryngeally modified consonants being different from ejectives "in
that the former have a different airstream mechanism from that of
the latter: ejectives are by definition glottalic egressive (with
glottalic egressive airstream mechanism, that is, with
outward/upward airflow where the body of air being moved is that
in the pharynx); the other segments can be assumed to be pulmonic
egressive (the sound is produced with (powered by) outward/upward
airflow where the body of air being moved is that in the lungs)"
(Shahin). "The mechanism that gets ejective airflow going,
involves trapping air between the closed vocal folds, closed
velum, and stop closure somewhere in the oral cavity. Reducing the
volume of this trapped air, by bobbing the larynx upwards,
increases the air pressure and on release of the stop closure air
flows out to equalise with atmospheric air. In contrast,
laryngealized/creaky/stiff voice is a quality achieved by
particular settings of the vocal folds (and associated structures)
using normal pulmonary airflow." (Reid) "This means that creaky
and stiff voice etc. are sub-categories of (or perhaps parallel
to) modal voice. Ejectives, on the other hand, are formed by the
closed larynx rising to increase air pressure behind the
occlusion. This airstream mechanism, the laryngeal or glottalic
airstream, is thus very different and cannot itself be mofified by
any laryngeal activity." (Jones) This explains why laryngealized
consonants may be voiced, whereas ejectives cannot.

Jessen speaks of the "distinct bursts" of ejectives, "[which] are
quite loud (strong burst amplitude) and are often isolated from
the following vowel through a long Voice Onset Time, esp. in North
American Indian languages, such as Navaho. Korean 'fortis' stops,
on the other hand, do not have particularly striking bursts. What
is most characteristic about them are the abrupt vowel onsets...
(stiff voice). In ejectives, on the other hand, voice quality
changes seem to be more variable."

He adds that glottal release may be realized by creaky voice. The
main function of the glottal release in ejectives seems to be the
higher pressure for a louder release of the consonant. The vowel
onset is later with ejectives. Laryngealized consonants have
glottal stop at the beginning, in the middle, or at the end of the
consonant. Implosives are often subsumed under the notion of
laryngealized consonants.

Menzel characterizes ejectives as a "voiceless stop-type closure
at *two* places, always including the glottis. Creaky voice, on
the other hand, are a voiced phenomenon, produced using a special
kind of vibration of the glottis... Creaky voice seems a kind of
"slowing down" of the vibration of the vocal cords. Now, voiced
stops have hardly any vibration because there is no time for it
and fricatives must create enough friction to be recognizable as
such, so you don't get the creaky voice effect with consonant,
only with vowels, liquids, and nasals.

Davis seems to be a very experienced researchers and hinted me to
the history of the glottalization discussion of the late
Sixties/early Seventies, which covered my question, too. He quotes
an area in Africa, where, interesting at least if we view areal
typology, nearby languages have either laryngealized or ejectives
consonants. Besides, he told me about gender variations in
American languages, the women of which had ejectives, the men of
which produced a "partial glottal closure", with the result of
voiced consonants (non-existent in women's speech).

According to Warner, ejectives involve creaky voicing at the
beginning of the following vowel in some languages. She prefers
the term "creaky voice" over the others.

Dace opposes the glottal closure of the ejectives to the glottal
vibration of the creaky voice.

Ronald Cosper describes the difference as follows: "Ejectives
involve total glottal closure, and the stopage occurs orally.
Creaky voice involves a partial opening of the glottis, with a
diminished voicing."

Peter Menzel reminds of the real counterpart of ejectives, which
are the implosives.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue