On Thu, 29 May 2014 12:17:55 +0300, Rusiok <[log in to unmask]> wrote:

>On 28/05/2014 19:20, Kev-Bill Walker wrote: > Really to give a sensible answer for a the additional letter, IMO we need to know how the other 26 are being used.
>  They are: a b c d e f g h i j k l m n o p q r s t u v w x y z ' a b c d e k w g a/h i/j ɟ k l ̥ /l m ̥ /m n ̥ /n o p o/ ʕ w r ̥ /r s t u/w ɡ w k w ʔ k ʔ c ʔ t ʔ e/ ʔ j

(You may want to try to format your mail for the list in a more plain-text friendly fashion.  The above is how your table came through to me.)

I take it your language is some sort of resurrected PIE, then.  How do your laryngeals <' h q> (or whatever) function phonologically?  When do they take their vocalic values, and when consonantal ones?  What is the difference between these and the non-alternating vowels <e a o>?  
If <' h q> can be regarded as parallel to <e a o>, then maybe the most _phonologically_ perspicuous notation would be accented versions of those letters, such as <e̯ a̯ o̯>.  I chose the inverted breve below diacritic there to suggest that these are really consonants despite the vowel letters.  Or if digraphs don't repel you, would <eh ah oh> or <he ha ho> be reasonable?

If you don't like those... your values <f v> /k_w g_w/ are kinda clever, but the glottalic stops do seem to have been shoved in wherever they fit.  So you could go to diacritics for the glottalic stops, use <q> for /?/ and move /?\/ to <x> or something.  

For what it's worth my own opinion on PIE /h_1 h_2 h_3/ is that they were [h X R].  Phonotactics and IIRC the Uralic borrowing evidence suggests that there probably wasn't a stop among them; /h_3/ wasn't relevantly rounded since the rounded obstruents that survive didn't cause /o/-colouring; and I'm partial to suggestions that the shape of the system of surviving dorsal stops results from earlier uvular stops, whereupon Ockham suggests sticking to uvulars rather than pharyngeals for these guys too.