Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!zombie.ncsc.mil!news.duke.edu!convex!cs.utexas.edu!howland.reston.ans.net!news.sprintlink.net!EU.net!uknet!festival!edcogsci!iad
From: iad@cogsci.ed.ac.uk (Ivan A Derzhanski)
Subject: Re: H and digraphs thereof
Message-ID: <CyADw0.ME9@cogsci.ed.ac.uk>
Organization: Centre for Cognitive Science, Edinburgh, UK
References: <CxyyKF.D5n@cogsci.ed.ac.uk> <CxzBt8.4J2@inter.NL.net>
Date: Wed, 26 Oct 1994 15:33:35 GMT
Lines: 37

In article <CxzBt8.4J2@inter.NL.net> mcv@inter.NL.net (Miguel Carrasquer) writes:
>In article <CxyyKF.D5n@cogsci.ed.ac.uk>, Ivan A Derzhanski <iad@cogsci.ed.ac.uk> wrote:
>>But _nh_ in Welsh is a voiceless _n_, not palatal or velar.  Yes, _ll_
>>is inconsistent.  (That's why JRR Tolkien uses _lh_ in his romanisation
>>for the Welsh-influenced Sindarin.)

In one of the appendices to _The Lord of the Rings_ JRR Tolkien
mentions unvoiced nasals `such as Old English _hn_ or Welsh _nh_'
(or something to that effect), and Jim Allan's essay on Sindarin in
_An Introduction to Elvish_ states that the Welsh sounds are just that.
The _ItE_ contributors sound very competent, and I don't doubt JRRT's
familiarity with Welsh (one of his two favourite languages).

>Well, so much for "Teach Yourself Welsh".  The Pronunciation chapter
>is completely silent on NH.

So are the pronunciation chapters of most Welsh grammars and textbooks.
I remember seeing one which did list _mh_, _nh_ and _ngh_ as separate
(unvoiced) consonants in our library, only I couldn't find it when I
looked for it yesterday.

What I did find is a book with the title _Welsh Phonology_, actually a
collection of articles edited by Martin J Ball, and an essay by him
and another by Glyn E Jones caught my eye.  What they say is that the
consonants in question are in fact clusters containing a nasal (unvoiced)
and an aspirated segment, thus [m<o>h], [n<o>h], [N<o>h], and they always
follow a homorganic voiced nasal; and GEJ further explains that they
can also be realised as [h{nsd}h], if the voicing ceases after the
oral closure for the nasal is realised rather than before it.

So this is what I can say on the subject.

-- 
`That's yer oan problem, Judas', they telt him.  `It's nae concern tae us.'
Ivan A Derzhanski (iad@cogsci.ed.ac.uk/chaos.cs.brandeis.edu)  (The G-- G--)
* Centre for Cognitive Science,  2 Buccleuch Place,   Edinburgh EH8 9LW,  UK
* Cowan House E113, Pollock Halls, 18 Holyrood Pk Rd, Edinburgh EH16 5BD, UK
