Re: aspirated m?

Ray Brown wrote:

>
> I find the notion of long (presumably geminate) voicless sonorants
> somewhat unlikely in a natlang.
Sure.  They'd been likely to become voiced.

> The trouble was we know was that JRRT never finished 'discovering'
> Sindarin or any other of his languages - he was always revising them and
> in those days when everything was papar-based it invevitably meant that
> some older bits hung around unrevised.
Which, as Tristan said, is probably not a bad thing!

>> It also appears that |ll|
>> is ambiguous between [l:] and [K] -- only the
>> etymology can tell which is which.
>
>       :)
Yes, the idea of determining stuff by etymology in
a conlang is a bit hilarious!

>
>> A bit strange, since Old English would have suggested
>> the spellings |hn, hñ, hm, hl| for the voiceless
>> sonorants.
>
>
> Yes - they are perhaps less ambiguous.
>
> BTW do we have any evidence that the Old English spells |hn|, |hl| and |hr|
>  did not indicate biphonemic groups, i.e. were pronounced something like
> [Xn], [Xl], {xr]?
We can't.  Alas the conventions of OE alliterations which
Tristan mentioned are not much to go by, since alliteration
patterns were higly conventionalized; thus any [g] could
alliterate with any [j], regardless if the [j] came from
Germanic */g/ or Germanic */j/!

In Icelandic |hr hl hn hj| are pronounced as unitary
voiceless sounds [r_0 K n_0 C|, but are still probably
biphonemic all the same!

--

/BP 8^)
--
Benct Philip Jonsson -- melroch at melroch dot se

         Solitudinem faciunt pacem appellant!
                                             (Tacitus)