On 3/23/2014 6:26 PM, Jörg Rhiemeier wrote:
> Hallo conlangers!
> On Sunday 23 March 2014 23:11:45 Herman Miller wrote:
>> On 3/23/2014 11:57 AM, Jörg Rhiemeier wrote:
>>> What is the current phoneme inventory of Tirëlat?  Is this page:
>>> still up to date or not?
>> It's pretty much accurate. Some of the romanization has changed. I use
>> double letters for long vowels instead of "h". I don't currently use [β]
>> as an allophone of /b/ between vowels. Borrowings from Jarda with /ł/
>> pronounced as [ɮ] (like the example of "kjełbu") are likely to change.
> Fine.  The script seems to place /ts dz/ and /ʂ ʐ/ in the same POA
> range, while grouping /s z/ with /t d/.  So the first may have been
> alveolar and the second dental, or the first postalveolar and the
> second alveolar.  Was this the case at some point in the history of
> the language?

I wouldn't read too much into the assignment of sounds to letters in the 
script. Most likely, the letters were used out of convenience when 
Tirëlat adopted the Kjaginits script, since modern Tirëlat lacks 
retroflex stops, and Kjaginits has no single letters for affricates.

But it is possible that the affricates could have been retroflex stops 
in early Tirëlat. But some Tirëlat dialects have two sets of affricates 
(alveolar and post-alveolar). So one possibility for early Tirëlat is 
that it had five points of articulation for stops.

p b
t̪ d̪
t d
ʈ ɖ
k ɡ

But in this scheme, it doesn't seem likely that the modern d/r sound 
([ð] vs. [ɾ] in different dialects) was another stop. It could have been 
a tap or some other kind of sound.

Or it could have been more like this:

p ɓ b
t ɗ d
ʈ ᶑ ɖ
c ʄ ɟ
k ɠ ɡ

where both the retroflex and palatal sets developed into affricates 
(except for the implosives). Following Alex Fink's suggestion that the 
palatal implosive developed into /j/, maybe something similar could have 
happened to the retroflex implosive, developing into something like /ɻ/. 
But there's not a sign of anything like /ɻ/ in the modern language.

I'm starting to think the affricates might have been affricates all 
along, instead of developing from stops...