LINGUIST List 8.1096

Sat Jul 26 1997

Sum: vowel deletion between like consonants

Editor for this issue: Martin Jacobsen <>


  1. LAS, summary: vowel deletion between two like consonants

Message 1: summary: vowel deletion between two like consonants

Date: Fri, 25 Jul 1997 15:59:49 -0400
From: LAS <>
Subject: summary: vowel deletion between two like consonants

Quite some time ago, I wrote requesting information concerning
instances of proper names where the pronunciation reflects a deletion
of a vowel occuring between two like consonants in the spelling of the
name. I cited such names as FARRER [fEr] WORCESTER [wUst^r] and
LEICESTER [lEst^r] and had hoped to find a whole slew of other names
exhibiting this phenomenon. Unfortunately, I got no responses which
dealt with proper nouns. However, I did get some interesting responses
discussing how this occurs in Hungarian, Arabic, and English outside
the realm of proper nouns. Thanks to all who responded. Sorry I didn't
post sooner. It's just that I had hoped to get something on proper
nouns before replying.

Sincerely, David Harris
David Harris
Language Analysis Systems Voice: (703) 834-6200 ext. 242
2214 Rock Hill Road, Suite 201 Fax: (703) 834-6230
Herndon, VA 22070 

Original Query:

>I've been thinking about names like Worcester, Leicester, and
>Gloucester which, though perhaps viewed by most of my fellow Americans
>as strange relics of British eccentricity, really do follow a simple
>and straightforward phonological process where two like consonantal
>continuants separated by a weak vowel (in this case schwa) are merged
>into one by the deletion of this weak vowel. The same phenomenon
>occurs with the name of a school in Provo, Utah which I attended as an

> 	Farrer Junior High School
> 	This is pronounced "Fair" (or "Ferr" if your accent makes a
> 	difference between [Er] and [eir] which mine doesn't)

> Anyway, using a regular expression, I searched the 1990 US Census
>list of 80,000 surnames available on the web in order to find names
>in which like consonants were separated by one vowel and came across
>a few more from various language groups that may or may not be
>examples of this. (It's hard to know when you can't have the names
>pronounced for you as well as see them written.) My question, then,
>is this: 
> I would be interested in knowing in what other languages this
>phenomenon occurs and with what other sounds. Judging from some of
>the names I encountered in my search, I suspect that it may occur in
>Japanese where [i] and [u] deletion causes two like consonants to be
>connected together. Also, in languages where word-internal gemination
>is phonemic, I'm curious as to whether simplification occurs as it
>does in English or if the gemination retains its full phonological
>value. Please include as many examples as you can with your comments
>and I will post a complete summary to the list. Any other
>observations about this phenomenon are also welcome.

Beginning of Responses:
1- Subject: Worcester
Here is a Hungarian example:

_ko2zta1rsasa1g_ /ko/starSASa:g/ fast/casual [ko/staSSa:g] `republic'
and another lexicalized one: _azt hiszem_ /Ast hisEm/ > [AsisEm] >
[assEm] `that-ACC believe-I' I cannot think of examples with
non-coronals, but that - I think _ has to do with statistics.

Symbols: o/ = IPA slashed o (round front mid vowel),
S = Eng sh
A = round back low vowel
E = IPA epsilon

Peter Szigetvari
2- From: (Dale Russell)
Subject: Phonological process--Worcester/Gloucester/Leicester

 > Also, in languages where word-internal gemination is phonemic, I'm
 > curious as to whether simplification occurs as it does in English
 > or if the gemination retains its full phonological value.

Does such simplification always occur in English? Does "meanness"
rhyme exactly with "Venus"? I've heard this example used to argue
that English does make at least limited use of the concept of the
mora, claiming that both /n/s in "meanness" get pronounced.

 Dale Russell
3- Subject: Deleted vowel

In response to your linguist posting: Note that Am. Eng. speakers
often delete the schwa between the /d/'s, so that "Where did he go"
becomes "Where'd he go", and "How did she do that" becomes "How'd she
do that".
- ------------------------------------------------------------------
Dan Loehr "Wherever you go, there you are."
Georgetown University - Buckaroo Bonzai
- --------------------------------------------------------------------------
4- From: "Robert Port" <>
Subject: deleted Vs in Eng
Date: Sun, 8 Sep 1996 21:45:59 -0500 (EST)
Cc: (Robert Port)

A student of mine wrote a PhD thesis on this topic many years ago. He
collected recordings of spoken Eng and looked for factors that would
account for why some unstressed Vs are deleted more than others in
words like parade, intelligent, invitation, etc. The thesis was
printed by the IU Linguistics Club. And I just happened to notice
that it is still available for a mere $3!
So check the web page at IU linguistics and you will find it in the
publications list of the IULC. Check it out:

Good luck. Bob Port

 ( ( ( O ) ) ) ( ( ( O ) ) ) ( ( ( O ) ) ) 
 Lingstcs/Comp Sci/Cogntv Sci
 ROBERT F. PORT 330 Memorial Hall, Indiana University
		 Bloomington, Indiana 47405
 812-855-9217 Fx 812-855-5363
5- dave,

your post reminded me of a related weak-vowel merger process that has
led to many (awkwardly) amusing misunderstandings. in my dialect
(southern californian US english), final -er/-or following Vr is
deleted, which, in combination with my vowel system, results in such
mergers as 'mirror'-->'mere', 'terror'-->'tear', 'error'-->'air', etc.
but the one that stands out (and which i've consciously had to change)
is 'horror', which, you guessed it, comes out 'whore'. imagine the,
uh, horror of telling non-californians that, as a kid, i used to watch
a lot of horror films! :-)


Gregory Ward
Department of Linguistics
Northwestern University
2016 Sheridan Road
Evanston IL 60208-4090

tel: 847-491-8055
fax: 847-491-3770

6- From:	Peter Daniels <>
Subject: first syllable of Farrer??
Content-Length: 503
Date:	Wed, 4 Sep 1996 21:25:24 -0500

I'm ore interested in your statement that the name Farrer could be
[fair(r)] or [fEr(r)]; i.e. = fairy or ferry. I realize that most
non-New Yorkers are seriously vowel-deprived in the low front region;
but I would naturally say [faerr] = farrow (something to do with
pigs, I think), same as in Harris.

Most New York Farrars, it seems, are final-stressed ([frar]), but the
publisher Farrar Straus & Giroux is initial-stressed, and in fact is
[faerr]. (We couldn't final-stress a Farrer, though.)

7- From:
Date: Thu, 5 Sep 1996 14:35:54 +0000
Subject: vowel elision
X-Pmrqc: 1
Priority: normal

Look into Arabic: the 3rd person singular of the perfect tense of
verbs is most commonly formed CaCaCa (C being the root consonants),
e.g. kataba "he wrote" The first person of this verb is katabtu "I

Now verbs with two identical consonants behave irregularly: while the
1st person sg. of the verb "to pass" is marartu "I passed", the 3rd
person is marra "he passed" instead of the expected *marara. The same
is true for many other Semitic languages (Hebrew etc.). 

Carsten Peust Seminar of Egyptology and Coptology Goettingen or
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue