THEORY: h huffnpuffery (was: RE: varia)

I hope the theory tag will excuse a further contribution to this thread.

Nik:
> Ed Heil wrote:
> > No, if there's *pharyngeal* turbulence, it's h-bar.  If there's
> > glottal turbulence, it's h.  If there's no turbulence at all, it's
> > silence.
>
> But it's hardly enough turbulence to call it a fricative!  It's merely a
> voiceless equivalent of the following vowel, for me.  It's physically
> impossible to have less turbulence, the vocal cords are as lax as
> possible, as lax as in any other voiceless sound.  I can produce a
> glottal fricative, and it's clearly different than /h/.
IIRC, Catford (my favourite textbook-writing phonetician) recognizes a
tripartite distinction between voiceless fricative, voiced fricative, and
approximant, where approximants are inherently voiced, their voiceless
counterparts being inaudible because there is insufficient constriction
to generate turbulence, which is the only way in which voiceless
continuants are audible. Hence [h] is a fricative. (It makes less sense
to try to categorize /h/.) Mind you, I recall being excoriated by Mark
Line for saying something along these lines three years or so ago, but
I cannot remember the ostensible reason.

ObConlang (somewhat drearily): [h] in Livagian is an allophone of /kh/
in almost all environments and also of /f/ and /th/ when adjacent to
a homorganic nasal or plosive. Hence _Ed Heil_ is /ed khail/ ([hail]
or [xail]) and _Nik Taylor_ is /nigkh dtheilr/, /gkh/ and /dth/ being [kh]
and [th] (or [kx] and [tT]) respectively.

--And.