Re: Underused phonemes

staving And Rosta:
>Peter Bleackley, On 13/01/2006 09:24:
>>It occurred to me yesterday that I don't know of any conlangs that have
>>[W] as a phoneme.
>
>You can't tell purely from a phoneme inventory whether [W] occurs in the
>conlang. For instance, Livagian lacks /W/, but has [W] as a realization of
>/hu/ -- i.e. like English (for those dialects that have [W]).
>
>>I was wondering what other sounds people might think worthy of more exposure.
>
>Somebody with spare time on their hands should do a Conlang poll that
>solicits and collates lists in CXS notation of all phones that occur
>contrastively in a given conlang. Ideally, the poll would be one of those
>automated online ones, else collating the results'd be murder (unless
>somebody writes a perl script or suchlike to do it). Alas, this exercise
>is beyond the limits of my competence and free time.
>
>Livagian, for example, has 24 segmental phonemes plus 3 tones, and 59
>mutually contrastive phones (factoring out tonal contrasts).
>
This reminds me, does anyone know of any software that, given the phonemic
inventory of a language, can calculate the minimum set of distinctive
features necessary to describe it? I'd like to use generative phonology to
create a descendent of Khangaþyagon (which I've previously described as
Wavoragon, but this may change).

Pete

(ps And - you have a Reply-to: header set to your personal address, rather
than the list)