LINGUIST List 14.1707

Mon Jun 16 2003

Sum: Uvular Fricatives and Glottal Stops

Editor for this issue: Karen Milligan <>


  1. Randall Gess, Summary: Uvulars/pharyngeals and glottal stop

Message 1: Summary: Uvulars/pharyngeals and glottal stop

Date: Mon, 16 Jun 2003 17:35:55 +0000
From: Randall Gess <>
Subject: Summary: Uvulars/pharyngeals and glottal stop

The following, belated summary is of responses to a query I sent to
Linguist List on May 1 (Linguist 14.1243), asking for synchronic or
diachronic alternations between uvular fricatives (or approximants)
and glottal stop. I would first like to thank those who responded to
my query, whose names appear below. The word limit prevents me from
listing other cases I have found myself, or from offering my proposed
explanation for these alternations. Anyone interested can email me and
I can send a handout from a recent talk that included discussion of
this alternation.

Mark Donohue suggests that in the Alor-Pantar languages the best
reconstruction for something that surfaces synchronically as [k], [x],
[h], [?] and zero in different languages and different positions, is
the voiceless velar stop with a uvular allophone.
In a more speculative case, Dominic Watt notes the possibility of an
alternation which might exist (or might have existed), between a
uvular fricative and glottal stop in some conservative, rural dialects
of Northumbrian English. Northumbrian English is somewhat famous for
the Northumbrian burr, a rhotic realized as a uvular fricative (or
approximant), now probably well on its way to extinction. In more
progressive dialects of Northumbrian (most particularly those of the
urban centers Newcastle upon Tyne, Gateshead, and Durham), where
rhoticity has disappeared, speakers have the option of inserting
*either* a linking [r] or a glottal stop in contexts where a word
historically ending in /r/ is followed by a vowel-initial word. So,
for example, on the Tyneside Metro (subway trains) youll hear taped
announcements saying either Stand clea[r] of the doors please or Stand
clea[?] of the doors please. Now, in this case the rhotic is an
alveolar or post-alveolar approximant. But Watt speculates that, in
the rural hinterland to the north of Tyneside, the alternation -
assuming the variety in question is now variably rhotic or fully
non-rhotic, *and* that it has retained the uvular fricative in spite
of the loss/attrition of rhoticity - could well be one where sequences
like car alarm can be pronounced either as ca[R] alarm or ca[?] alarm.

Several respondents to my Linguist List query pointed to developments
of pharyngeals and epiglottals to glottal stop.

According to Martin H�mmel, in some Sardic dialects there is a change
of [k] to glottal stop, and that in neighboring dialects there is a
parallel change of [k] to a fricative which varies, according to
Continis (1987) instrumental study, between a velar, uvular and
epiglottal articulation. According to K�mmel, Contini states
explicitly that this fricative is a lenition product of /k/, and that
the glottal stop in the other dialects is a secondary development of
the fricative. In the fricative dialects, the change occurs in
postvocalic and intervocalic position, while in the glottal stop
dialects, the novel segment also occurs in absolute initial position,
this possibly also a secondary generalization. H�mmel further reports
that in some Semitic languages (i.e., Akkadian, (Neo-)Aramaic, New
South Arabian Soqotri, dialects of Tigrinya, a number of South
Ethiopian languages (especially Amharic), and also in some varieties
of modern Hebrew), there is a development from a voiced uvular
fricative to a pharyngeal (or epiglottal) fricative that later becomes
a glottal stop. (The Aramaic shift was also pointed out to me by
Robert Hoberman, although he states that the shift involved both
voiced and voiceless uvulars.)

Ghil`ad Zuckermann points out that in Modern Hebrew, the original
Hebrew voiced pharyngeal fricative is normally not realized, but is
realized as glottal stop when in postconsonantal position in uncommon
words. In addition, the original Hebrew voiceless pharyngeal is
realized in Modern Hebrew as a uvular fricative.

Kimary Shahin states that in Arabic the voiced pharyngeal fricative
(which is probably actually an approximant) is sometimes output as
glottal stop.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue