Re: signal and noise in phonologies and scripts

Anton Sherwood <bronto@...> wrote:

> Jörg Rhiemeier wrote:
> > Something to write into Danny Wier's guestbook ;-)
>
> hein?
Danny Wier is a list member who has the habit of posting yet another
version of the phoneme inventory of his conlang Tech every few months.
Apparently, he constantly revises the phonology and hardly gets
beyond that.  The point is that his phoneme inventories are always
*monstrous*, with more than 100 consonants, employing lots of
secondary articulations, glottalization, aspiration and other features
which would make all those phonemes hard to distinguish from each other.

He rarely comments on other issues.

> > >
> > >         |__|
> > >            |__|
> > >
> > > with analytic functions of
> > >
> > >          \______
> > >          /
> > >         /
> > >
> > > in the complex plane. [mathematical stuff snipped]
> >
> > What does the last figure represent?
>
> It has the same topology and lengths as the first figure, but with the
> bends straightened (and the angles equalized; but that's not so
> important).
Yes, it has, but yet it is virtually unrecognizable; and there are lots
of examples of letters that are topologically equivalent and yet
different in real-world scripts.  For example, the letters C, G, I, J,
L, M, N, S, U, V, W, Z are topologically equivalent in many sans-serif
fonts in consisting of a single, non-branching line each that is merely
bent in different ways.

There is more to a letter shape than topology and relative lengths of
the legs; bends and turns (and in many scripts, spatial orientation)
are just as important, as the examples given
above demonstrate.

But it is of course upon you what your script looks like, of course.

Jörg.

.

Reply