Re: aspirated m?

On Wednesday, November 24, 2004, at 12:08 , Rene Uittenbogaard wrote:

[snip]
> I understand now how |mh| is pronounced in Gaelic and Brittonic
> languages. I just hope I won't come across the term "aspirated m" on
> some conlang page :-/
You can never tell with a conlang   ;)

>> (Also: the Gaelic & welsh sounds written |mh| are pronounced *very
>> differently* and come from quite different origins)
>
> To sum up:
> Gaelic: |m| -> (soft mutation)    |mh| /v/ or /w/;
Yes.

> Welsh:  |p| -> (nasal mutation)   |mh| /m_h/;
Nasal mutation - yes. But not /m_h/; phomemically it is /mh/ amd
pronounced like /mh/ in "somehow".

I think I implied in my earlier mail that nasal mutation occurs also in
Breton & Cornish. That was an error - it does not. Besides the 'soft
mutation', Cornish and Breton have the spirant (as Welsh does), a 'hard
mutation' and a 'mixed mutation'.

> and incidentally:
> Welsh:  |m| -> (soft mutation)    |f| /v/;
Yes.

> Welsh:  |p| -> (spirant mutation) |f| /v/
No. The spirant mutation affects only the consonants |p| /p/, |t| /t/ and
|c| /k/. It is shoen in writing by adding |h| and in each case it is
question of fricativization in pronounciation, thus:
|p| --> |ph| /f/
|t| --> |th| /T/
|c| --> |ch| /X/

Worth noting that in Welsh spelling:
/v/ is always spelled |f|
/f/ is spelled |ph| if it is the spirant mutation of |p|, but otherwise as
|ff|.

> Wow, thanks a lot, Ray! That was very enlightening.
You're welcome.
================================================

On Tuesday, November 23, 2004, at 09:37 , Isaac Penzev wrote:

> Ray Brown eskribiw:
>
>
>> Indeed, while voiceless /m/ is not uncommon as an allophone, as for
>> example in English _small_, I do not know of any natural language in
>> which
>> voiced & voiceless nasals are separate phonemes. I am _not_ making a
>> dogmatic statement that they do not exist - that would be foolish - I am
>> merely saying I do not know of any.
>
> AFAIK, voiceless nasals are separate phonemes in Burmese (Myanmar).
Ah, Burmese - the language which, like English, eschews phonemic spelling
in favor of etymological spelling  :)

Now you mention it, I have heard this. But following this up on the
Internet has so far yielded very inconclusive data. Some sites even speak
of 'aspirated nasals'.    :)

It does seem certain that either [m_0], [n_0] and [N_0] or [m_h], [n_h]
and [N_h] occur in Burmese, but it is less clear that they are phonemic.
Indeed one site stated quite categorically that the voiceless nasals are
in complementary distribution with their voiced counterparts and are,
therefore, _not_ phonemic.

===============================================

On Wednesday, November 24, 2004, at 04:41 , Kris Kowal wrote:

> --- Steven Williams:
>> --- Ray Brown eskribiw:
>> --- Isaac Penzev skrev:
>>> AFAIK, voiceless nasals are separate phonemes in
>>> Burmese (Myanmar).
>> And several Native American languages,
A wee bit vague IMO. There are, I understand, something approaching 1000
Native American languages between the Arctic Circle and Tierra del Fuego.
Which particular languages are you referring to. What examples are there
of the _phonemic_ status of the voiceless nasals?

>> not to mention
>> at least one of Tolkien's conlangs (if that counts).
Conlangs cannot count since - as it has often been observed - anything is
possible in a conlang. I am very aware that JRRT did strive for
'naturalism' in his artlangs, but the quy was not infallible - tho I must
admit that so far I have discovered nothing implausible in the Tolklangs.

> Huh?  Either you know something about Telerin or Ilkorin that I don't,
> but I don't believe that any of Tolkien's conlangs have voicles
> nasals.
Quite - I've (rather hastily I admit) gone through all my notes on
Tolklangs & find no reference.

> I only really know for sure in Quenya and Sindarin.  However,
> there are tenwa that could concievably be used for the purpose.
Yes, yes - in any schematic system like tenwa there is likely to the
possibility of signs for sounds that do not occur or even could not occur
because they are physically impossible.
===============================================

On Wednesday, November 24, 2004, at 06:32 , Andreas Johansson wrote:
[snip]
> FWIW, in the Etymologies, there is one revision layer that changes
> Noldorin
> m->sm- to hm-, apparently denoting a voiceless m.
Did JRRT specifically say that |hm| was a voiceless _m_? Without such
testimony, we rely on guesswork. There are other ways hm- could be
pronounced.

A similar change is thought to have occurred in Proto-Greek, sm- -->
hm- --> m-. It is incontrovertible that initial s- before vowels became h-.

We also know that initial sm- of Proto-Greek becomes just m- ancient Greek,
  tho occasionally one or two older forms did survive long enough to get
into writing in some dialects, before disappearing, like _smikros_ (small)
  besides the far more common and eventually standard _mikros_. It is
therefore thought that the change was sm- --> hm- --> m-

But where -sm- occurred medially the result is -mm- which Michel Lejeune
suggests developed thus: -sm- --> -zm- --> -h\m- --> -mm-

(I should note that initial sm- did occur in Ancient Greek in words of
foreign origin and -sm-did occur medially as the result of later sound
changes)

There is no evidence that [m_0] or [m_h] occurred _phonemically_ at any
time in Greek - tho I guess [m_0] is the pronunciation of /m/ in words
like _tmesis_ in both ancient & modern Greek.

I know well enough that [m_0] occurs in natlangs; after all it occurs in
English _small_ and Welsh _ysmygu_ (to smoke). But, despite the replies so
far, I am still left in the position of not knowing of any natlangs where
it is can clearly be shown that voiceless nasals have phonemic status.

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com
===============================================
Anything is possible in the fabulous Celtic twilight,
which is not so much a twilight of the gods
as of the reason."      [JRRT, "English and Welsh" ]

Reply