Re: Phonological Oddity

[andrew <hobbit@mail.earthlight.co.nz>]

On Sun, 1 Aug 1999, Geoff Eddy wrote:

> I'm sending this particular message to you privately because there seems
> to be some sort of problem with celticonlang which drops several of my
> postings. I'm not sure what's going on, but this way is certain to reach
> you.
>
I was away from this address for two months last year and I cancelled or
suspended all my subscriptions during that time.  Celticonlang was one of
the lists I have not been able to find information on to resubscribe.
Asking on the Brithenig mailing list I now send to Celticonlang, at least
it no longer bounces back, but I receive no postings from it.  If you have
any information on subscribing to this group I would be grateful to
receive it.
 
> - In Irish, /nt/ reduces to /d/ both internally and as one instance of
> eclipsis (nasal mutation). E.g. "cent" for 100 becomes _cead_.
> 
> - In Breathanach (and Liotan), the same happens (but only in LF): 100 is
> _cead_ too.
> 
> - In Welsh, it's not quite so clear, but we have _canu_ for "to sing",
> compared to Latin _cantare_, with the medial /n/ corresponding to the
> /nh/ which would result from nasal mutation.
> 
> - But in Brithenig, we have _cantar_, preserving the stop medially,
> whereas /t/ undergoes nasal mutation to /nh/ initially, as in Welsh.
> 
> Put another way, if lenition (soft mutation) operates both medially
> between vowels and initially after a word ending in a vowel, is there a
> reason why clusters of nasal + stop undergo nasal mutation initially,
> but not medially?
> 
> 
> I hope this makes sense! Could you let me know if you've already seen
> this through celticonlang? Perhaps the fact it's literally travelling to
> the other side of the world has something to do with it.
>
I've given you a description of my position in relation to celticonlang
above, but you have picked up one of the dirty little secrets of
Brithenig: medial nasal mutation is tricky and confusing.  In Welsh I have
read that unvoiced nasal clusters become Nh before the stress, and then N
after it.  It seems to me that looking through what dictionaries I have
access to, that clusters were later introduced and I have no idea at what
point this happened.  I considered it simpler to quietly ignore this
feature and worry about it at some other time.  So it's probably time for
me to crank up the revision engines and revisit this feature.

- andrew.
--
Andrew Smith, Intheologus 			hobbit@earthlight.co.nz

	Lo! thy dread empire, Chaos! is restored;
	Light dies before thy uncreating word:
	Thy hand, great Anarch! lets the curtain fall;
	And Universal Darkness buries All.
			- Alexander Pope, The Dunciad, Book IV.