Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site spar.UUCP
Path: utzoo!watmath!clyde!cbosgd!ihnp4!qantel!hplabs!pesnta!greipa!decwrl!spar!ellis
From: ellis@spar.UUCP (Michael Ellis)
Newsgroups: net.nlang
Subject: Re: Pet Peeves
Message-ID: <644@spar.UUCP>
Date: Fri, 8-Nov-85 09:40:56 EST
Article-I.D.: spar.644
Posted: Fri Nov  8 09:40:56 1985
Date-Received: Mon, 11-Nov-85 06:14:59 EST
References: <747@cyb-eng.UUCP> <1900005@datacube.UUCP> <2260@iddic.UUCP> <670@stc-b.stc.UUCP> <216@dcdwest.UUCP> <131@daisy.UUCP>
Reply-To: ellis@spar.UUCP (Michael Ellis)
Organization: Schlumberger Palo Alto Research, CA
Lines: 30

>How about "nuclear" pronounced "noo-que-lur", as if spelled "noocuulur"?
> - David Schachter

     Phonetically, this is a simple swapping of two adjacent consonants:

	nuclear /nuwkl0y0r/	 vs 	nucular? /nuwky0l0r/

     ..which puts it in the same category as errors like
     
        relevant /rel0v0nt/	vs	revelant? /rev0l0nt/

     I believe this phonetic error is called `metathesis'.

     Other errors mentioned seem to be due to English's natural process
     of weakening unaccented vowels and elision of unpronounceable
     clusters of glides [rlwy]:
   
	library /laybrer1/ => /laybr0r1/ => /laybrr1 = laybr1/ libry?
	february /febr0wer1/ => /febr0w0r1/ => /febrwr1,febr1/ febry?

     BTW, the pronunciation `febuary' is not really the dropping of an `r';
     rather, it is the substition of /y/ for /r/:

	february /febr0wer1/	vs	febuary? /feby0wer1/

     There are many historical instances of repeated glides within a word
     dissimilating in this way, even when separated by intervening consonants.
     Purple (Lat. purpur-) and marble (Lat. marmor-), for example.

-michael